-<!-- $Id$ --!>\r
-<html>\r
-\r
-<head>\r
-<meta http-equiv="Content-Type"\r
-content="text/html; charset=iso-8859-1">\r
-<title>Computing Health Expectancies using IMaCh</title>\r
-<html>\r
-\r
-<head>\r
-<meta http-equiv="Content-Type"\r
-content="text/html; charset=iso-8859-1">\r
-<title>IMaCh</title>\r
-</head>\r
-\r
-<body bgcolor="#FFFFFF">\r
-\r
-<hr size="3" color="#EC5E5E">\r
-\r
-<h1 align="center"><font color="#00006A">Computing Health\r
-Expectancies using IMaCh</font></h1>\r
-\r
-<h1 align="center"><font color="#00006A" size="5">(a Maximum\r
-Likelihood Computer Program using Interpolation of Markov Chains)</font></h1>\r
-\r
-<p align="center"> </p>\r
-\r
-<p align="center"><a href="http://www.ined.fr/"><img\r
-src="logo-ined.gif" border="0" width="151" height="76"></a><img\r
-src="euroreves2.gif" width="151" height="75"></p>\r
-\r
-<h3 align="center"><a href="http://www.ined.fr/"><font\r
-color="#00006A">INED</font></a><font color="#00006A"> and </font><a\r
-href="http://euroreves.ined.fr"><font color="#00006A">EUROREVES</font></a></h3>\r
-\r
-<p align="center"><font color="#00006A" size="4"><strong>Version\r
-0.97, June 2004</strong></font></p>\r
-\r
-<hr size="3" color="#EC5E5E">\r
-\r
-<p align="center"><font color="#00006A"><strong>Authors of the\r
-program: </strong></font><a href="http://sauvy.ined.fr/brouard"><font\r
-color="#00006A"><strong>Nicolas Brouard</strong></font></a><font\r
-color="#00006A"><strong>, senior researcher at the </strong></font><a\r
-href="http://www.ined.fr"><font color="#00006A"><strong>Institut\r
-National d'Etudes Démographiques</strong></font></a><font\r
-color="#00006A"><strong> (INED, Paris) in the "Mortality,\r
-Health and Epidemiology" Research Unit </strong></font></p>\r
-\r
-<p align="center"><font color="#00006A"><strong>and Agnès\r
-Lièvre<br clear="left">\r
-</strong></font></p>\r
-\r
-<h4><font color="#00006A">Contribution to the mathematics: C. R.\r
-Heathcote </font><font color="#00006A" size="2">(Australian\r
-National University, Canberra).</font></h4>\r
-\r
-<h4><font color="#00006A">Contact: Agnès Lièvre (</font><a\r
-href="mailto:lievre@ined.fr"><font color="#00006A"><i>lievre@ined.fr</i></font></a><font\r
-color="#00006A">) </font></h4>\r
-\r
-<hr>\r
-\r
-<ul>\r
- <li><a href="#intro">Introduction</a> </li>\r
- <li><a href="#data">On what kind of data can it be used?</a></li>\r
- <li><a href="#datafile">The data file</a> </li>\r
- <li><a href="#biaspar">The parameter file</a> </li>\r
- <li><a href="#running">Running Imach</a> </li>\r
- <li><a href="#output">Output files and graphs</a> </li>\r
- <li><a href="#example">Exemple</a> </li>\r
-</ul>\r
-\r
-<hr>\r
-\r
-<h2><a name="intro"><font color="#00006A">Introduction</font></a></h2>\r
-\r
-<p>This program computes <b>Healthy Life Expectancies</b> from <b>cross-longitudinal\r
-data</b> using the methodology pioneered by Laditka and Wolf (1).\r
-Within the family of Health Expectancies (HE), Disability-free\r
-life expectancy (DFLE) is probably the most important index to\r
-monitor. In low mortality countries, there is a fear that when\r
-mortality declines, the increase in DFLE is not proportionate to\r
-the increase in total Life expectancy. This case is called the <em>Expansion\r
-of morbidity</em>. Most of the data collected today, in\r
-particular by the international <a href="http://www.reves.org">REVES</a>\r
-network on Health expectancy, and most HE indices based on these\r
-data, are <em>cross-sectional</em>. It means that the information\r
-collected comes from a single cross-sectional survey: people from\r
-various ages (but mostly old people) are surveyed on their health\r
-status at a single date. Proportion of people disabled at each\r
-age, can then be measured at that date. This age-specific\r
-prevalence curve is then used to distinguish, within the\r
-stationary population (which, by definition, is the life table\r
-estimated from the vital statistics on mortality at the same\r
-date), the disable population from the disability-free\r
-population. Life expectancy (LE) (or total population divided by\r
-the yearly number of births or deaths of this stationary\r
-population) is then decomposed into DFLE and DLE. This method of\r
-computing HE is usually called the Sullivan method (from the name\r
-of the author who first described it).</p>\r
-\r
-<p>Age-specific proportions of people disabled (prevalence of\r
-disability) are dependent on the historical flows from entering\r
-disability and recovering in the past until today. The age-specific\r
-forces (or incidence rates), estimated over a recent period of time\r
-(like for period forces of mortality), of entering disability or\r
-recovering a good health, are reflecting current conditions and\r
-therefore can be used at each age to forecast the future of this\r
-cohort<em>if nothing changes in the future</em>, i.e to forecast the\r
-prevalence of disability of each cohort. Our finding (2) is that the period\r
-prevalence of disability (computed from period incidences) is lower\r
-than the cross-sectional prevalence. For example if a country is\r
-improving its technology of prosthesis, the incidence of recovering\r
-the ability to walk will be higher at each (old) age, but the\r
-prevalence of disability will only slightly reflect an improve because\r
-the prevalence is mostly affected by the history of the cohort and not\r
-by recent period effects. To measure the period improvement we have to\r
-simulate the future of a cohort of new-borns entering or leaving at\r
-each age the disability state or dying according to the incidence\r
-rates measured today on different cohorts. The proportion of people\r
-disabled at each age in this simulated cohort will be much lower that\r
-the proportions observed at each age in a cross-sectional survey. This\r
-new prevalence curve introduced in a life table will give a more\r
-realistic HE level than the Sullivan method which mostly measured the\r
-History of health conditions in this country.</p>\r
-\r
-<p>Therefore, the main question is how to measure incidence rates\r
-from cross-longitudinal surveys? This is the goal of the IMaCH\r
-program. From your data and using IMaCH you can estimate period\r
-HE and not only Sullivan's HE. Also the standard errors of the HE\r
-are computed.</p>\r
-\r
-<p>A cross-longitudinal survey consists in a first survey\r
-("cross") where individuals from different ages are\r
-interviewed on their health status or degree of disability. At\r
-least a second wave of interviews ("longitudinal")\r
-should measure each new individual health status. Health\r
-expectancies are computed from the transitions observed between\r
-waves and are computed for each degree of severity of disability\r
-(number of life states). More degrees you consider, more time is\r
-necessary to reach the Maximum Likelihood of the parameters\r
-involved in the model. Considering only two states of disability\r
-(disable and healthy) is generally enough but the computer\r
-program works also with more health statuses.<br>\r
-<br>\r
-The simplest model is the multinomial logistic model where <i>pij</i>\r
-is the probability to be observed in state <i>j</i> at the second\r
-wave conditional to be observed in state <em>i</em> at the first\r
-wave. Therefore a simple model is: log<em>(pij/pii)= aij +\r
-bij*age+ cij*sex,</em> where '<i>age</i>' is age and '<i>sex</i>'\r
-is a covariate. The advantage that this computer program claims,\r
-comes from that if the delay between waves is not identical for\r
-each individual, or if some individual missed an interview, the\r
-information is not rounded or lost, but taken into account using\r
-an interpolation or extrapolation. <i>hPijx</i> is the\r
-probability to be observed in state <i>i</i> at age <i>x+h</i>\r
-conditional to the observed state <i>i</i> at age <i>x</i>. The\r
-delay '<i>h</i>' can be split into an exact number (<i>nh*stepm</i>)\r
-of unobserved intermediate states. This elementary transition (by\r
-month or quarter trimester, semester or year) is modeled as a\r
-multinomial logistic. The <i>hPx</i> matrix is simply the matrix\r
-product of <i>nh*stepm</i> elementary matrices and the\r
-contribution of each individual to the likelihood is simply <i>hPijx</i>.\r
-<br>\r
-</p>\r
-\r
-<p>The program presented in this manual is a quite general\r
-program named <strong>IMaCh</strong> (for <strong>I</strong>nterpolated\r
-<strong>MA</strong>rkov <strong>CH</strong>ain), designed to\r
-analyse transition data from longitudinal surveys. The first step\r
-is the parameters estimation of a transition probabilities model\r
-between an initial status and a final status. From there, the\r
-computer program produces some indicators such as observed and\r
-stationary prevalence, life expectancies and their variances and\r
-graphs. Our transition model consists in absorbing and\r
-non-absorbing states with the possibility of return across the\r
-non-absorbing states. The main advantage of this package,\r
-compared to other programs for the analysis of transition data\r
-(For example: Proc Catmod of SAS<sup>®</sup>) is that the whole\r
-individual information is used even if an interview is missing, a\r
-status or a date is unknown or when the delay between waves is\r
-not identical for each individual. The program can be executed\r
-according to parameters: selection of a sub-sample, number of\r
-absorbing and non-absorbing states, number of waves taken in\r
-account (the user inputs the first and the last interview), a\r
-tolerance level for the maximization function, the periodicity of\r
-the transitions (we can compute annual, quarterly or monthly\r
-transitions), covariates in the model. It works on Windows or on\r
-Unix.<br>\r
-</p>\r
-\r
-<hr>\r
-\r
-<p>(1) Laditka, Sarah B. and Wolf, Douglas A. (1998), "New\r
-Methods for Analyzing Active Life Expectancy". <i>Journal of\r
-Aging and Health</i>. Vol 10, No. 2. </p>\r
-<p>(2) <a href=http://taylorandfrancis.metapress.com/app/home/contribution.asp?wasp=1f99bwtvmk5yrb7hlhw3&referrer=parent&backto=issue,1,2;journal,2,5;linkingpublicationresults,1:300265,1\r
->Lièvre A., Brouard N. and Heathcote Ch. (2003) Estimating Health Expectancies \r
-from Cross-longitudinal surveys. <em>Mathematical Population Studies</em>.- 10(4), pp. 211-248</a>\r
-\r
-<hr>\r
-\r
-<h2><a name="data"><font color="#00006A">On what kind of data can\r
-it be used?</font></a></h2>\r
-\r
-<p>The minimum data required for a transition model is the\r
-recording of a set of individuals interviewed at a first date and\r
-interviewed again at least one another time. From the\r
-observations of an individual, we obtain a follow-up over time of\r
-the occurrence of a specific event. In this documentation, the\r
-event is related to health status at older ages, but the program\r
-can be applied on a lot of longitudinal studies in different\r
-contexts. To build the data file explained into the next section,\r
-you must have the month and year of each interview and the\r
-corresponding health status. But in order to get age, date of\r
-birth (month and year) is required (missing values is allowed for\r
-month). Date of death (month and year) is an important\r
-information also required if the individual is dead. Shorter\r
-steps (i.e. a month) will more closely take into account the\r
-survival time after the last interview.</p>\r
-\r
-<hr>\r
-\r
-<h2><a name="datafile"><font color="#00006A">The data file</font></a></h2>\r
-\r
-<p>In this example, 8,000 people have been interviewed in a\r
-cross-longitudinal survey of 4 waves (1984, 1986, 1988, 1990). Some\r
-people missed 1, 2 or 3 interviews. Health statuses are healthy (1)\r
-and disable (2). The survey is not a real one. It is a simulation of\r
-the American Longitudinal Survey on Aging. The disability state is\r
-defined if the individual missed one of four ADL (Activity of daily\r
-living, like bathing, eating, walking). Therefore, even if the\r
-individuals interviewed in the sample are virtual, the information\r
-brought with this sample is close to the situation of the United\r
-States. Sex is not recorded is this sample. The LSOA survey is biased\r
-in the sense that people living in an institution were not surveyed at\r
-first pass in 1984. Thus the prevalence of disability in 1984 is\r
-biased downwards at old ages. But when people left their household to\r
-an institution, they have been surveyed in their institution in 1986,\r
-1988 or 1990. Thus incidences are not biased. But cross-sectional\r
-prevalences of disability at old ages are thus artificially increasing\r
-in 1986, 1988 and 1990 because of a higher weight of people\r
-institutionalized in the sample. Our article shows the\r
-opposite: the period prevalence is lower at old ages than the\r
-adjusted cross-sectional prevalence proving important current progress\r
-against disability.</p>\r
-\r
-<p>Each line of the data set (named <a href="data1.txt">data1.txt</a>\r
-in this first example) is an individual record. Fields are separated\r
-by blanks: </p>\r
-\r
-<ul>\r
- <li><b>Index number</b>: positive number (field 1) </li>\r
- <li><b>First covariate</b> positive number (field 2) </li>\r
- <li><b>Second covariate</b> positive number (field 3) </li>\r
- <li><a name="Weight"><b>Weight</b></a>: positive number\r
- (field 4) . In most surveys individuals are weighted\r
- according to the stratification of the sample.</li>\r
- <li><b>Date of birth</b>: coded as mm/yyyy. Missing dates are\r
- coded as 99/9999 (field 5) </li>\r
- <li><b>Date of death</b>: coded as mm/yyyy. Missing dates are\r
- coded as 99/9999 (field 6) </li>\r
- <li><b>Date of first interview</b>: coded as mm/yyyy. Missing\r
- dates are coded as 99/9999 (field 7) </li>\r
- <li><b>Status at first interview</b>: positive number.\r
- Missing values ar coded -1. (field 8) </li>\r
- <li><b>Date of second interview</b>: coded as mm/yyyy.\r
- Missing dates are coded as 99/9999 (field 9) </li>\r
- <li><strong>Status at second interview</strong> positive\r
- number. Missing values ar coded -1. (field 10) </li>\r
- <li><b>Date of third interview</b>: coded as mm/yyyy. Missing\r
- dates are coded as 99/9999 (field 11) </li>\r
- <li><strong>Status at third interview</strong> positive\r
- number. Missing values ar coded -1. (field 12) </li>\r
- <li><b>Date of fourth interview</b>: coded as mm/yyyy.\r
- Missing dates are coded as 99/9999 (field 13) </li>\r
- <li><strong>Status at fourth interview</strong> positive\r
- number. Missing values are coded -1. (field 14) </li>\r
- <li>etc</li>\r
-</ul>\r
-\r
-<p> </p>\r
-\r
-<p>If your longitudinal survey does not include information about\r
-weights or covariates, you must fill the column with a number\r
-(e.g. 1) because a missing field is not allowed.</p>\r
-\r
-<hr>\r
-\r
-<h2><font color="#00006A">Your first example parameter file</font><a\r
-href="http://euroreves.ined.fr/imach"></a><a name="uio"></a></h2>\r
-\r
-<h2><a name="biaspar"></a>#Imach version 0.97b, June 2004,\r
-INED-EUROREVES </h2>\r
-\r
-<p>This first line was a comment. Comments line start with a '#'.</p>\r
-\r
-<h4><font color="#FF0000">First uncommented line</font></h4>\r
-\r
-<pre>title=1st_example datafile=data1.txt lastobs=8600 firstpass=1 lastpass=4</pre>\r
-\r
-<ul>\r
- <li><b>title=</b> 1st_example is title of the run. </li>\r
- <li><b>datafile=</b> data1.txt is the name of the data set.\r
- Our example is a six years follow-up survey. It consists\r
- in a baseline followed by 3 reinterviews. </li>\r
- <li><b>lastobs=</b> 8600 the program is able to run on a\r
- subsample where the last observation number is lastobs.\r
- It can be set a bigger number than the real number of\r
- observations (e.g. 100000). In this example, maximisation\r
- will be done on the 8600 first records. </li>\r
- <li><b>firstpass=1</b> , <b>lastpass=4 </b>In case of more\r
- than two interviews in the survey, the program can be run\r
- on selected transitions periods. firstpass=1 means the\r
- first interview included in the calculation is the\r
- baseline survey. lastpass=4 means that the information\r
- brought by the 4th interview is taken into account.</li>\r
-</ul>\r
-\r
-<p> </p>\r
-\r
-<h4><a name="biaspar-2"><font color="#FF0000">Second uncommented\r
-line</font></a></h4>\r
-\r
-<pre>ftol=1.e-08 stepm=1 ncovcol=2 nlstate=2 ndeath=1 maxwav=4 mle=1 weight=0</pre>\r
-\r
-<ul>\r
- <li><b>ftol=1e-8</b> Convergence tolerance on the function\r
- value in the maximisation of the likelihood. Choosing a\r
- correct value for ftol is difficult. 1e-8 is a correct\r
- value for a 32 bits computer.</li>\r
- <li><b>stepm=1</b> Time unit in months for interpolation.\r
- Examples:<ul>\r
- <li>If stepm=1, the unit is a month </li>\r
- <li>If stepm=4, the unit is a trimester</li>\r
- <li>If stepm=12, the unit is a year </li>\r
- <li>If stepm=24, the unit is two years</li>\r
- <li>... </li>\r
- </ul>\r
- </li>\r
- <li><b>ncovcol=2</b> Number of covariate columns included in the\r
- datafile before the column of the date of birth. You can have\r
-covariates that won't necessary be used during the\r
- run. It is not the number of covariates that will be\r
- specified by the model. The 'model' syntax describes the\r
- covariates to be taken into account during the run. </li>\r
- <li><b>nlstate=2</b> Number of non-absorbing (alive) states.\r
- Here we have two alive states: disability-free is coded 1\r
- and disability is coded 2. </li>\r
- <li><b>ndeath=1</b> Number of absorbing states. The absorbing\r
- state death is coded 3. </li>\r
- <li><b>maxwav=4</b> Number of waves in the datafile.</li>\r
- <li><a name="mle"><b>mle</b></a><b>=1</b> Option for the\r
- Maximisation Likelihood Estimation. <ul>\r
- <li>If mle=1 the program does the maximisation and\r
- the calculation of health expectancies </li>\r
- <li>If mle=0 the program only does the calculation of\r
- the health expectancies and other indices and graphs\r
-but without the maximization.. </li>\r
- There also other possible values:\r
- <ul>\r
- <li>If mle=-1 you get a template which can be useful if\r
-your model is complex with many covariates.</li>\r
- <li> If mle=-3 IMaCh computes the mortality but without\r
- any health status (May 2004)</li> <li>If mle=2 IMach\r
- likelihood corresponds to a linear interpolation</li> <li>\r
- If mle=3 IMach likelihood corresponds to an exponential\r
- inter-extrapolation</li> \r
- <li> If mle=4 IMach likelihood\r
- corresponds to no inter-extrapolation, and thus biasing\r
- the results. </li> \r
- <li> If mle=5 IMach likelihood\r
- corresponds to no inter-extrapolation, and before the\r
- correction of the Jackson's bug (avoid this).</li>\r
- </ul>\r
- </ul>\r
- </li>\r
- <li><b>weight=0</b> Possibility to add weights. <ul>\r
- <li>If weight=0 no weights are included </li>\r
- <li>If weight=1 the maximisation integrates the\r
- weights which are in field <a href="#Weight">4</a></li>\r
- </ul>\r
- </li>\r
-</ul>\r
-\r
-<h4><font color="#FF0000">Covariates</font></h4>\r
-\r
-<p>Intercept and age are systematically included in the model.\r
-Additional covariates can be included with the command: </p>\r
-\r
-<pre>model=<em>list of covariates</em></pre>\r
-\r
-<ul>\r
- <li>if<strong> model=. </strong>then no covariates are\r
- included</li>\r
- <li>if <strong>model=V1</strong> the model includes the first\r
- covariate (field 2)</li>\r
- <li>if <strong>model=V2 </strong>the model includes the\r
- second covariate (field 3)</li>\r
- <li>if <strong>model=V1+V2 </strong>the model includes the\r
- first and the second covariate (fields 2 and 3)</li>\r
- <li>if <strong>model=V1*V2 </strong>the model includes the\r
- product of the first and the second covariate (fields 2\r
- and 3)</li>\r
- <li>if <strong>model=V1+V1*age</strong> the model includes\r
- the product covariate*age</li>\r
-</ul>\r
-\r
-<p>In this example, we have two covariates in the data file\r
-(fields 2 and 3). The number of covariates included in the data\r
-file between the id and the date of birth is ncovcol=2 (it was\r
-named ncov in version prior to 0.8). If you have 3 covariates in\r
-the datafile (fields 2, 3 and 4), you will set ncovcol=3. Then\r
-you can run the programme with a new parametrisation taking into\r
-account the third covariate. For example, <strong>model=V1+V3 </strong>estimates\r
-a model with the first and third covariates. More complicated\r
-models can be used, but it will takes more time to converge. With\r
-a simple model (no covariates), the programme estimates 8\r
-parameters. Adding covariates increases the number of parameters\r
-: 12 for <strong>model=V1, </strong>16 for <strong>model=V1+V1*age\r
-</strong>and 20 for <strong>model=V1+V2+V3.</strong></p>\r
-\r
-<h4><font color="#FF0000">Guess values for optimization</font><font\r
-color="#00006A"> </font></h4>\r
-\r
-<p>You must write the initial guess values of the parameters for\r
-optimization. The number of parameters, <em>N</em> depends on the\r
-number of absorbing states and non-absorbing states and on the\r
-number of covariates. <br>\r
-<em>N</em> is given by the formula <em>N</em>=(<em>nlstate</em> +\r
-<em>ndeath</em>-1)*<em>nlstate</em>*<em>ncovmodel</em> . <br>\r
-<br>\r
-Thus in the simple case with 2 covariates (the model is log\r
-(pij/pii) = aij + bij * age where intercept and age are the two\r
-covariates), and 2 health degrees (1 for disability-free and 2\r
-for disability) and 1 absorbing state (3), you must enter 8\r
-initials values, a12, b12, a13, b13, a21, b21, a23, b23. You can\r
-start with zeros as in this example, but if you have a more\r
-precise set (for example from an earlier run) you can enter it\r
-and it will speed up them<br>\r
-Each of the four lines starts with indices "ij": <b>ij\r
-aij bij</b> </p>\r
-\r
-<blockquote>\r
- <pre># Guess values of aij and bij in log (pij/pii) = aij + bij * age\r
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">\r
+<!-- saved from url=(0044)http://euroreves.ined.fr/imach/doc/imach.htm -->\r
+<!-- $Id$ --!><HTML><HEAD><TITLE>Computing Health Expectancies using IMaCh</TITLE>\r
+<META content="text/html; charset=iso-8859-1" http-equiv=Content-Type>\r
+<META content="MSHTML 5.00.3315.2870" name=GENERATOR></HEAD>\r
+<BODY bgColor=#ffffff>\r
+<HR color=#ec5e5e SIZE=3> \r
+\r
+<H1 align=center><FONT color=#00006a>Computing Health Expectancies using \r
+IMaCh</FONT></H1>\r
+<H1 align=center><FONT color=#00006a size=5>(a Maximum Likelihood Computer \r
+Program using Interpolation of Markov Chains)</FONT></H1>\r
+<P align=center> </P>\r
+<P align=center><A href="http://www.ined.fr/"><IMG border=0 height=76 \r
+src="Computing Health Expectancies using IMaCh_fichiers/logo-ined.gif" \r
+width=151></A><IMG height=75 \r
+src="Computing Health Expectancies using IMaCh_fichiers/euroreves2.gif" \r
+width=151></P>\r
+<H3 align=center><A href="http://www.ined.fr/"><FONT \r
+color=#00006a>INED</FONT></A><FONT color=#00006a> and </FONT><A \r
+href="http://euroreves.ined.fr/"><FONT color=#00006a>EUROREVES</FONT></A></H3>\r
+<P align=center><FONT color=#00006a size=4><STRONG>Version 0.97, June \r
+2004</STRONG></FONT></P>\r
+<HR color=#ec5e5e SIZE=3>\r
+\r
+<P align=center><FONT color=#00006a><STRONG>Authors of the program: \r
+</STRONG></FONT><A href="http://sauvy.ined.fr/brouard"><FONT \r
+color=#00006a><STRONG>Nicolas Brouard</STRONG></FONT></A><FONT \r
+color=#00006a><STRONG>, senior researcher at the </STRONG></FONT><A \r
+href="http://www.ined.fr/"><FONT color=#00006a><STRONG>Institut National \r
+d'Etudes Démographiques</STRONG></FONT></A><FONT color=#00006a><STRONG> (INED, \r
+Paris) in the "Mortality, Health and Epidemiology" Research Unit \r
+</STRONG></FONT></P>\r
+<P align=center><FONT color=#00006a><STRONG>and Agnès Lièvre<BR \r
+clear=left></STRONG></FONT></P>\r
+<H4><FONT color=#00006a>Contribution to the mathematics: C. R. Heathcote \r
+</FONT><FONT color=#00006a size=2>(Australian National University, \r
+Canberra).</FONT></H4>\r
+<H4><FONT color=#00006a>Contact: Agnès Lièvre (</FONT><A \r
+href="mailto:lievre@ined.fr"><FONT \r
+color=#00006a><I>lievre@ined.fr</I></FONT></A><FONT color=#00006a>) </FONT></H4>\r
+<HR>\r
+\r
+<UL>\r
+ <LI><A \r
+ href="http://euroreves.ined.fr/imach/doc/imach.htm#intro">Introduction</A> \r
+ <LI><A href="http://euroreves.ined.fr/imach/doc/imach.htm#data">On what kind \r
+ of data can it be used?</A> \r
+ <LI><A href="http://euroreves.ined.fr/imach/doc/imach.htm#datafile">The data \r
+ file</A> \r
+ <LI><A href="http://euroreves.ined.fr/imach/doc/imach.htm#biaspar">The \r
+ parameter file</A> \r
+ <LI><A href="http://euroreves.ined.fr/imach/doc/imach.htm#running">Running \r
+ Imach</A> \r
+ <LI><A href="http://euroreves.ined.fr/imach/doc/imach.htm#output">Output files \r
+ and graphs</A> \r
+ <LI><A href="http://euroreves.ined.fr/imach/doc/imach.htm#example">Exemple</A> \r
+ </LI></UL>\r
+<HR>\r
+\r
+<H2><A name=intro><FONT color=#00006a>Introduction</FONT></A></H2>\r
+<P>This program computes <B>Healthy Life Expectancies</B> from \r
+<B>cross-longitudinal data</B> using the methodology pioneered by Laditka and \r
+Wolf (1). Within the family of Health Expectancies (HE), disability-free life \r
+expectancy (DFLE) is probably the most important index to monitor. In low \r
+mortality countries, there is a fear that when mortality declines (and therefore total life expectancy improves), the increase will not be as great, leading to an <EM>Expansion of morbidity</EM>. Most of the data collected today, \r
+in particular by the international <A href="http://www.reves.org/">REVES</A> \r
+network on Health Expectancy and the disability process, and most HE indices based on these data, are \r
+<EM>cross-sectional</EM>. This means that the information collected comes from a \r
+single cross-sectional survey: people from a variety of ages (but often old people) \r
+are surveyed on their health status at a single date. The proportion of people \r
+disabled at each age can then be estimated at that date. This age-specific \r
+prevalence curve is used to distinguish, within the stationary population \r
+(which, by definition, is the life table estimated from the vital statistics on \r
+mortality at the same date), the disabled population from the disability-free \r
+population. Life expectancy (LE) (or total population divided by the yearly \r
+number of births or deaths of this stationary population) is then decomposed \r
+into disability-free life expectancy (DFLE) and disability life \r
+expectancy (DLE). This method of computing HE is usually called the Sullivan \r
+method (after the author who first described it).</P>\r
+<P>The age-specific proportions of people disabled (prevalence of disability) are \r
+dependent upon the historical flows from entering disability and recovering in the past. The age-specific forces (or incidence rates) of entering \r
+disability or recovering a good health, estimated over a recent period of time (as period forces of mortality), are reflecting current conditions and \r
+therefore can be used at each age to forecast the future of this cohort <EM>if \r
+nothing changes in the future</EM>, i.e to forecast the prevalence of disability of each cohort. Our finding (2) is that the period prevalence of disability \r
+(computed from period incidences) is lower than the cross-sectional prevalence. \r
+For example if a country is improving its technology of prosthesis, the \r
+incidence of recovering the ability to walk will be higher at each (old) age, \r
+but the prevalence of disability will only slightly reflect an improvement because \r
+the prevalence is mostly affected by the history of the cohort and not by recent \r
+period effects. To measure the period improvement we have to simulate the future \r
+of a cohort of new-borns entering or leaving the disability state or \r
+dying at each age according to the incidence rates measured today on different cohorts. The \r
+proportion of people disabled at each age in this simulated cohort will be much \r
+lower that the proportions observed at each age in a cross-sectional survey. \r
+This new prevalence curve introduced in a life table will give a more realistic \r
+HE level than the Sullivan method which mostly reflects the history of health \r
+conditions in a country.</P>\r
+<P>Therefore, the main question is how can we measure incidence rates from \r
+cross-longitudinal surveys? This is the goal of the IMaCH program. From your \r
+data and using IMaCH you can estimate period HE as well as the Sullivan HE. In addition the standard errors of the HE are computed.</P>\r
+<P>A cross-longitudinal survey consists of a first survey ("cross") where \r
+individuals of different ages are interviewed about their health status or degree \r
+of disability. At least a second wave of interviews ("longitudinal") should \r
+measure each individual new health status. Health expectancies are computed from \r
+the transitions observed between waves (interviews) and are computed for each degree of \r
+severity of disability (number of health states). The more degrees of severity considered, the more \r
+time is necessary to reach the Maximum Likelihood of the parameters involved in \r
+the model. Considering only two states of disability (disabled and healthy) is \r
+generally enough but the computer program works also with more health \r
+states.<BR><BR>The simplest model for the transition probabilities is the multinomial logistic model where \r
+<I>pij</I> is the probability to be observed in state <I>j</I> at the second \r
+wave conditional to be observed in state <EM>i</EM> at the first wave. Therefore \r
+a simple model is: log<EM>(pij/pii)= aij + bij*age+ cij*sex,</EM> where \r
+'<I>age</I>' is age and '<I>sex</I>' is a covariate. The advantage that this \r
+computer program claims, is that if the delay between waves is not \r
+identical for each individual, or if some individual missed an interview, the \r
+information is not rounded or lost, but taken into account using an \r
+interpolation or extrapolation. <I>hPijx</I> is the probability to be observed \r
+in state <I>i</I> at age <I>x+h</I> conditional on the observed state <I>i</I> \r
+at age <I>x</I>. The delay '<I>h</I>' can be split into an exact number \r
+(<I>nh*stepm</I>) of unobserved intermediate states. This elementary transition \r
+(by month or quarter, trimester, semester or year) is modeled as the above multinomial \r
+logistic. The <I>hPx</I> matrix is simply the matrix product of <I>nh*stepm</I> \r
+elementary matrices and the contribution of each individual to the likelihood is \r
+simply <I>hPijx</I>. <BR></P>\r
+<P>The program presented in this manual is a general program named \r
+<STRONG>IMaCh</STRONG> (for <STRONG>I</STRONG>nterpolated \r
+<STRONG>MA</STRONG>rkov <STRONG>CH</STRONG>ain), designed to analyse transitions from longitudinal surveys. The first step is the estimation of the set of the parameters of a model for the \r
+transition probabilities between an initial state and a final state. \r
+From there, the computer program produces indicators such as the observed and \r
+stationary prevalence, life expectancies and their variances both numerically and graphically. Our \r
+transition model consists of absorbing and non-absorbing states assuming the \r
+possibility of return across the non-absorbing states. The main advantage of \r
+this package, compared to other programs for the analysis of transition data \r
+(for example: Proc Catmod of SAS<SUP>®</SUP>) is that the whole individual \r
+information is used even if an interview is missing, a state or a date is \r
+unknown or when the delay between waves is not identical for each individual. \r
+The program is dependent upon a set of parameters inputted by the user: selection of a sub-sample, \r
+number of absorbing and non-absorbing states, number of waves to be taken in account , a tolerance level for the \r
+maximization function, the periodicity of the transitions (we can compute \r
+annual, quarterly or monthly transitions), covariates in the model. IMaCh works on \r
+Windows or on Unix platform.<BR></P>\r
+<HR>\r
+\r
+<P>(1) Laditka S. B. and Wolf, D. (1998), New Methods for Analyzing \r
+Active Life Expectancy. <I>Journal of Aging and Health</I>. Vol 10, No. 2. </P>\r
+<P>(2) <A \r
+href="http://taylorandfrancis.metapress.com/app/home/contribution.asp?wasp=1f99bwtvmk5yrb7hlhw3&referrer=parent&backto=issue,1,2;journal,2,5;linkingpublicationresults,1:300265,1">Lièvre \r
+A., Brouard N. and Heathcote Ch. (2003) Estimating Health Expectancies from \r
+Cross-longitudinal surveys. <EM>Mathematical Population Studies</EM>.- 10(4), \r
+pp. 211-248</A> \r
+<HR>\r
+\r
+<H2><A name=data><FONT color=#00006a>What kind of data is required?</FONT></A></H2>\r
+<P>The minimum data required for a transition model is the recording of a set of \r
+individuals interviewed at a first date and interviewed once more. From the observations of an individual, we obtain a follow-up over \r
+time of the occurrence of a specific event. In this documentation, the event is \r
+related to health state, but the program can be applied to many\r
+longitudinal studies with different contexts. To build the data file\r
+as explained \r
+in the next section, you must have the month and year of each interview and \r
+the corresponding health state. In order to get age, date of birth (month \r
+and year) are required (missing values are allowed for month). Date of death \r
+(month and year) is an important information also required if the individual is \r
+dead. Shorter steps (i.e. a month) will more closely take into account the \r
+survival time after the last interview.</P>\r
+<HR>\r
+\r
+<H2><A name=datafile><FONT color=#00006a>The data file</FONT></A></H2>\r
+<P>In this example, 8,000 people have been interviewed in a cross-longitudinal \r
+survey of 4 waves (1984, 1986, 1988, 1990). Some people missed 1, 2 or 3 \r
+interviews. Health states are healthy (1) and disabled (2). The survey is not a \r
+real one but a simulation of the American Longitudinal Survey on Aging. The \r
+disability state is defined as dependence in at least one of four ADLs (Activities \r
+of daily living, like bathing, eating, walking). Therefore, even if the \r
+individuals interviewed in the sample are virtual, the information in \r
+this sample is close to reality for the United States. Sex is not recorded \r
+is this sample. The LSOA survey is biased in the sense that people\r
+living in an institution were not included in the first interview in\r
+1984. Thus the prevalence of disability observed in 1984 is lower than\r
+the true prevalence at old ages. However when people moved into an\r
+institution, they were interviewed there in 1986, 1988 and 1990. Thus\r
+the incidences of disabilities are not biased. Cross-sectional \r
+prevalences of disability at old ages are thus artificially increasing in 1986, \r
+1988 and 1990 because of a greater proportion of the sample\r
+institutionalized. Our article (Lièvre A., Brouard N. and Heathcote\r
+Ch. (2003)) shows the opposite: the period prevalence based on the\r
+incidences is lower at old \r
+ages than the adjusted cross-sectional prevalence illustrating that\r
+there has been significant progress against disability.</P>\r
+<P>Each line of the data set (named <A \r
+href="http://euroreves.ined.fr/imach/doc/data1.txt">data1.txt</A> in this first \r
+example) is an individual record. Fields are separated by blanks: </P>\r
+<UL>\r
+ <LI><B>Index number</B>: positive number (field 1) \r
+ <LI><B>First covariate</B> positive number (field 2) \r
+ <LI><B>Second covariate</B> positive number (field 3) \r
+ <LI><A name=Weight><B>Weight</B></A>: positive number (field 4) . In most \r
+ surveys individuals are weighted to account for stratification of the \r
+ sample. \r
+ <LI><B>Date of birth</B>: coded as mm/yyyy. Missing dates are coded as 99/9999 \r
+ (field 5) \r
+ <LI><B>Date of death</B>: coded as mm/yyyy. Missing dates are coded as 99/9999 \r
+ (field 6) \r
+ <LI><B>Date of first interview</B>: coded as mm/yyyy. Missing dates are coded \r
+ as 99/9999 (field 7) \r
+ <LI><B>Status at first interview</B>: positive number. Missing values ar coded \r
+ -1. (field 8) \r
+ <LI><B>Date of second interview</B>: coded as mm/yyyy. Missing dates are coded \r
+ as 99/9999 (field 9) \r
+ <LI><STRONG>Status at second interview</STRONG> positive number. Missing \r
+ values ar coded -1. (field 10) \r
+ <LI><B>Date of third interview</B>: coded as mm/yyyy. Missing dates are coded \r
+ as 99/9999 (field 11) \r
+ <LI><STRONG>Status at third interview</STRONG> positive number. Missing values \r
+ ar coded -1. (field 12) \r
+ <LI><B>Date of fourth interview</B>: coded as mm/yyyy. Missing dates are coded \r
+ as 99/9999 (field 13) \r
+ <LI><STRONG>Status at fourth interview</STRONG> positive number. Missing \r
+ values are coded -1. (field 14) \r
+ <LI>etc </LI></UL>\r
+<P> </P>\r
+<P>If you do not wish to include information on weights or \r
+covariates, you must fill the column with a number (e.g. 1) since all\r
+fields must be present.</P>\r
+<HR>\r
+\r
+<H2><FONT color=#00006a>Your first example parameter file</FONT><A \r
+href="http://euroreves.ined.fr/imach"></A><A name=uio></A></H2>\r
+<H2><A name=biaspar></A>#Imach version 0.97b, June 2004, INED-EUROREVES </H2>\r
+<P>This first line was a comment. Comments line start with a '#'.</P>\r
+<H4><FONT color=#ff0000>First uncommented line</FONT></H4><PRE>title=1st_example datafile=data1.txt lastobs=8600 firstpass=1 lastpass=4</PRE>\r
+<UL>\r
+ <LI><B>title=</B> 1st_example is title of the run. \r
+ <LI><B>datafile=</B> data1.txt is the name of the data set. Our example is a \r
+ six years follow-up survey. It consists of a baseline followed by 3 \r
+ reinterviews. \r
+ <LI><B>lastobs=</B> 8600 the program is able to run on a subsample where the \r
+ last observation number is lastobs. It can be set a bigger number than the \r
+ real number of observations (e.g. 100000). In this example, maximisation will \r
+ be done on the first 8600 records. \r
+ <LI><B>firstpass=1</B> , <B>lastpass=4 </B>If there are more than two interviews \r
+ in the survey, the program can be run on selected transitions periods. \r
+ firstpass=1 means the first interview included in the calculation is the \r
+ baseline survey. lastpass=4 means that the last interview to be\r
+ included will be by the 4th. </LI></UL>\r
+<P> </P>\r
+<H4><A name=biaspar-2><FONT color=#ff0000>Second uncommented \r
+line</FONT></A></H4><PRE>ftol=1.e-08 stepm=1 ncovcol=2 nlstate=2 ndeath=1 maxwav=4 mle=1 weight=0</PRE>\r
+<UL>\r
+ <LI><B>ftol=1e-8</B> Convergence tolerance on the function value in the \r
+ maximisation of the likelihood. Choosing a correct value for ftol is \r
+ difficult. 1e-8 is the correct value for a 32 bit computer. \r
+ <LI><B>stepm=1</B> The time unit in months for interpolation. Examples:\r
+ <UL>\r
+ <LI>If stepm=1, the unit is a month \r
+ <LI>If stepm=4, the unit is a trimester \r
+ <LI>If stepm=12, the unit is a year \r
+ <LI>If stepm=24, the unit is two years \r
+ <LI>... </LI></UL>\r
+ <LI><B>ncovcol=2</B> Number of covariate columns included in the datafile \r
+ before the column for the date of birth. You can include covariates\r
+ that will not be used in the model as this number is not the number of covariates that will \r
+ be specified by the model. The 'model' syntax describes the covariates to be \r
+ taken into account during the run. \r
+ <LI><B>nlstate=2</B> Number of non-absorbing (alive) states. Here we have two \r
+ alive states: disability-free is coded 1 and disability is coded 2. \r
+ <LI><B>ndeath=1</B> Number of absorbing states. The absorbing state death is \r
+ coded 3. \r
+ <LI><B>maxwav=4</B> Number of waves in the datafile. \r
+ <LI><A name=mle><B>mle</B></A><B>=1</B> Option for the Maximisation Likelihood \r
+ Estimation. \r
+ <UL>\r
+ <LI>If mle=1 the program does the maximisation and the calculation of health \r
+ expectancies \r
+ <LI>If mle=0 the program only does the calculation of the health \r
+ expectancies and other indices and graphs but without the maximization. \r
+ There are also other possible values: \r
+ <UL>\r
+ <LI>If mle=-1 you get a template for the number of parameters\r
+ and the size of the variance-covariance matrix. This is useful if the model is \r
+ complex with many covariates. \r
+ <LI>If mle=-3 IMaCh computes the mortality but without any health status \r
+ (May 2004) \r
+ <LI>If mle=2 IMach likelihood corresponds to a linear interpolation \r
+ <LI>If mle=3 IMach likelihood corresponds to an exponential \r
+ inter-extrapolation \r
+ <LI>If mle=4 IMach likelihood corresponds to no inter-extrapolation, thus biasing the results. \r
+ <LI>If mle=5 IMach likelihood corresponds to no inter-extrapolation, and \r
+ before the correction of the Jackson's bug (avoid this). </LI></UL></LI></UL>\r
+ <LI><B>weight=0</B> Provides the possibility of adding weights. \r
+ <UL>\r
+ <LI>If weight=0 no weights are included \r
+ <LI>If weight=1 the maximisation integrates the weights which are in field \r
+ <A href="http://euroreves.ined.fr/imach/doc/imach.htm#Weight">4</A> \r
+ </LI></UL></LI></UL>\r
+<H4><FONT color=#ff0000>Covariates</FONT></H4>\r
+<P>Intercept and age are automatically included in the model. Additional \r
+covariates can be included with the command: </P><PRE>model=<EM>list of covariates</EM></PRE>\r
+<UL>\r
+ <LI>if<STRONG> model=. </STRONG>then no covariates are included \r
+ <LI>if <STRONG>model=V1</STRONG> the model includes the first covariate (field \r
+ 2) \r
+ <LI>if <STRONG>model=V2 </STRONG>the model includes the second covariate \r
+ (field 3) \r
+ <LI>if <STRONG>model=V1+V2 </STRONG>the model includes the first and the \r
+ second covariate (fields 2 and 3) \r
+ <LI>if <STRONG>model=V1*V2 </STRONG>the model includes the product of the \r
+ first and the second covariate (fields 2 and 3) \r
+ <LI>if <STRONG>model=V1+V1*age</STRONG> the model includes the product \r
+ covariate*age </LI></UL>\r
+<P>In this example, we have two covariates in the data file (fields 2 and 3). \r
+The number of covariates included in the data file between the id and the date \r
+of birth is ncovcol=2 (it was named ncov in version prior to 0.8). If you have 3 \r
+covariates in the datafile (fields 2, 3 and 4), you will set ncovcol=3. Then you \r
+can run the programme with a new parametrisation taking into account the third \r
+covariate. For example, <STRONG>model=V1+V3 </STRONG>estimates a model with the \r
+first and third covariates. More complicated models can be used, but this will \r
+take more time to converge. With a simple model (no covariates), the programme \r
+estimates 8 parameters. Adding covariates increases the number of parameters : \r
+12 for <STRONG>model=V1, </STRONG>16 for <STRONG>model=V1+V1*age </STRONG>and 20 \r
+for <STRONG>model=V1+V2+V3.</STRONG></P>\r
+<H4><FONT color=#ff0000>Guess values for optimization</FONT><FONT color=#00006a> \r
+</FONT></H4>\r
+<P>You must write the initial guess values of the parameters for optimization. \r
+The number of parameters, <EM>N</EM> depends on the number of absorbing states \r
+and non-absorbing states and on the number of covariates in the model (ncovmodel). <BR><EM>N</EM> is \r
+given by the formula <EM>N</EM>=(<EM>nlstate</EM> + \r
+<EM>ndeath</EM>-1)*<EM>nlstate</EM>*<EM>ncovmodel</EM> . <BR><BR>Thus in \r
+the simple case with 2 covariates in the model(the model is log (pij/pii) = aij + bij * age \r
+where intercept and age are the two covariates), and 2 health states (1 for \r
+disability-free and 2 for disability) and 1 absorbing state (3), you must enter \r
+8 initials values, a12, b12, a13, b13, a21, b21, a23, b23. You can start with \r
+zeros as in this example, but if you have a more precise set (for example from \r
+an earlier run) you can enter it and it will speed up the convergence<BR>Each of the four \r
+lines starts with indices "ij": <B>ij aij bij</B> </P>\r
+<BLOCKQUOTE><PRE># Guess values of aij and bij in log (pij/pii) = aij + bij * age\r
12 -14.155633 0.110794 \r
13 -7.925360 0.032091 \r
21 -1.890135 -0.029473 \r
-23 -6.234642 0.022315 </pre>\r
-</blockquote>\r
-\r
-<p>or, to simplify (in most of cases it converges but there is no\r
-warranty!): </p>\r
-\r
-<blockquote>\r
- <pre>12 0.0 0.0\r
+23 -6.234642 0.022315 </PRE></BLOCKQUOTE>\r
+<P>or, to simplify (in most of cases it converges but there is no warranty!): \r
+</P>\r
+<BLOCKQUOTE><PRE>12 0.0 0.0\r
13 0.0 0.0\r
21 0.0 0.0\r
-23 0.0 0.0</pre>\r
-</blockquote>\r
-\r
-<p>In order to speed up the convergence you can make a first run\r
-with a large stepm i.e stepm=12 or 24 and then decrease the stepm\r
-until stepm=1 month. If newstepm is the new shorter stepm and\r
-stepm can be expressed as a multiple of newstepm, like newstepm=n\r
-stepm, then the following approximation holds: </p>\r
-\r
-<pre>aij(stepm) = aij(n . stepm) - ln(n)\r
-</pre>\r
-\r
-<p>and </p>\r
-\r
-<pre>bij(stepm) = bij(n . stepm) .</pre>\r
-\r
-<p>For example if you already ran for a 6 months interval and\r
-got:<br>\r
-</p>\r
-\r
-<pre># Parameters\r
+23 0.0 0.0</PRE></BLOCKQUOTE>\r
+<P>In order to speed up the convergence you can make a first run with a large \r
+stepm i.e stepm=12 or 24 and then decrease the stepm until stepm=1 month. If \r
+newstepm is the new shorter stepm and stepm can be expressed as a multiple of \r
+newstepm, like newstepm=n stepm, then the following approximation holds: </P><PRE>aij(stepm) = aij(n . stepm) - ln(n)\r
+</PRE>\r
+<P>and </P><PRE>bij(stepm) = bij(n . stepm) .</PRE>\r
+<P>For example if you already ran with stepm=6 (a 6 months interval) and got:<BR></P><PRE># Parameters\r
12 -13.390179 0.126133 \r
13 -7.493460 0.048069 \r
21 0.575975 -0.041322 \r
23 -4.748678 0.030626 \r
-</pre>\r
-\r
-<p>If you now want to get the monthly estimates, you can guess\r
-the aij by substracting ln(6)= 1,7917<br>\r
-and running<br>\r
-</p>\r
-\r
-<pre>12 -15.18193847 0.126133 \r
+</PRE>\r
+<P>Then you now want to get the monthly estimates, you can guess the aij by \r
+subtracting ln(6)= 1.7917<BR>and running using<BR></P><PRE>12 -15.18193847 0.126133 \r
13 -9.285219469 0.048069\r
21 -1.215784469 -0.041322\r
23 -6.540437469 0.030626\r
-</pre>\r
-\r
-<p>and get<br>\r
-</p>\r
-\r
-<pre>12 -15.029768 0.124347 \r
+</PRE>\r
+<P>and get<BR></P><PRE>12 -15.029768 0.124347 \r
13 -8.472981 0.036599 \r
21 -1.472527 -0.038394 \r
23 -6.553602 0.029856 \r
\r
-which is closer to the results. The approximation is probably useful\r
+<P>which is closer to the results. The approximation is probably useful\r
only for very small intervals and we don't have enough experience to\r
-know if you will speed up the convergence or not.\r
-</pre>\r
-\r
-<pre> -ln(12)= -2.484\r
+know if you will speed up the convergence or not.<BR></P>\r
+</PRE><PRE> -ln(12)= -2.484\r
-ln(6/1)=-ln(6)= -1.791\r
-ln(3/1)=-ln(3)= -1.0986\r
-ln(12/6)=-ln(2)= -0.693\r
-</pre>\r
-\r
-In version 0.9 and higher you can still have valuable results even if\r
-your stepm parameter is bigger than a month. The idea is to run with\r
-bigger stepm in order to have a quicker convergence at the price of a\r
-small bias. Once you know which model you want to fit, you can put\r
-stepm=1 and wait hours or days to get the convergence!\r
-\r
-To get unbiased results even with large stepm we introduce the idea of\r
-pseudo likelihood by interpolating two exact likelihoods. Let us\r
-detail this:\r
-<p>\r
-If the interval of <em>d</em> months between two waves is not a\r
-mutliple of 'stepm', but is comprised between <em>(n-1) stepm</em> and\r
-<em>n stepm</em> then both exact likelihoods are computed (the\r
-contribution to the likelihood at <em>n stepm</em> requires one matrix\r
-product more) (let us remember that we are modelling the probability\r
-to be observed in a particular state after <em>d</em> months being\r
-observed at a particular state at 0). The distance, (<em>bh</em> in\r
-the program), from the month of interview to the rounded date of <em>n\r
-stepm</em> is computed. It can be negative (interview occurs before\r
-<em>n stepm</em>) or positive if the interview occurs after <em>n\r
-stepm</em> (and before <em>(n+1)stepm</em>).\r
-<br>\r
-Then the final contribution to the total likelihood is a weighted\r
-average of these two exact likelihoods at <em>n stepm</em> (out) and\r
-at <em>(n-1)stepm</em>(savm). We did not want to compute the third\r
-likelihood at <em>(n+1)stepm</em> because it is too costly in time, so\r
-we used an extrapolation if <em>bh</em> is positive. <br> Formula of\r
-inter/extrapolation may vary according to the value of parameter mle:\r
-<pre>\r
-mle=1 lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */\r
-\r
-mle=2 lli= (savm[s1][s2]>(double)1.e-8 ? \\r
+</PRE>In version 0.9 and higher you can still have valuable results even if your \r
+stepm parameter is bigger than a month. The idea is to run with bigger stepm in \r
+order to have a quicker convergence at the price of a small bias. Once you know \r
+which model you want to fit, you can put stepm=1 and wait hours or days to get \r
+the convergence! To get unbiased results even with large stepm we introduce the \r
+idea of pseudo likelihood by interpolating two exact likelihoods. In\r
+more detail: \r
+<P>If the interval of <EM>d</EM> months between two waves is not a multiple of \r
+'stepm', but is between <EM>(n-1) stepm</EM> and <EM>n stepm</EM> then \r
+both exact likelihoods are computed (the contribution to the likelihood at <EM>n \r
+stepm</EM> requires one matrix product more) (let us remember that we are \r
+modelling the probability to be observed in a particular state after <EM>d</EM> \r
+months being observed at a particular state at 0). The distance, (<EM>bh</EM> in \r
+the program), from the month of interview to the rounded date of <EM>n \r
+stepm</EM> is computed. It can be negative (interview occurs before <EM>n \r
+stepm</EM>) or positive if the interview occurs after <EM>n stepm</EM> (and \r
+before <EM>(n+1)stepm</EM>). <BR>Then the final contribution to the total \r
+likelihood is a weighted average of these two exact likelihoods at <EM>n \r
+stepm</EM> (out) and at <EM>(n-1)stepm</EM>(savm). We did not want to compute \r
+the third likelihood at <EM>(n+1)stepm</EM> because it is too costly in time, so \r
+we used an extrapolation if <EM>bh</EM> is positive. <BR>The formula\r
+for the inter/extrapolation may vary according to the value of parameter mle: <PRE>mle=1 lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */\r
+ \r
+mle=2 lli= (savm[s1][s2]>(double)1.e-8 ? \\r
log((1.+bbh)*out[s1][s2]- bbh*(savm[s1][s2])): \\r
log((1.+bbh)*out[s1][s2])); /* linear interpolation */\r
-mle=3 lli= (savm[s1][s2]>1.e-8 ? \\r
+mle=3 lli= (savm[s1][s2]>1.e-8 ? \\r
(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]): \\r
log((1.+bbh)*out[s1][s2])); /* exponential inter-extrapolation */\r
\r
mle=4 lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* No interpolation */\r
no need to save previous likelihood into memory.\r
-</pre>\r
-<p>\r
-If the death occurs between first and second pass, and for example\r
-more precisely between <em>n stepm</em> and <em>(n+1)stepm</em> the\r
-contribution of this people to the likelihood is simply the difference\r
-between the probability of dying before <em>n stepm</em> and the\r
-probability of dying before <em>(n+1)stepm</em>. There was a bug in\r
-version 0.8 and death was treated as any other state, i.e. as if it\r
-was an observed death at second pass. This was not precise but\r
-correct, but when information on the precise month of death came\r
-(death occuring prior to second pass) we did not change the likelihood\r
-accordingly. Thanks to Chris Jackson for correcting us. In earlier\r
+</PRE>\r
+<P>If the death occurs between the first and second pass, and for example more \r
+precisely between <EM>n stepm</EM> and <EM>(n+1)stepm</EM> the contribution of \r
+these people to the likelihood is simply the difference between the probability \r
+of dying before <EM>n stepm</EM> and the probability of dying before \r
+<EM>(n+1)stepm</EM>. There was a bug in version 0.8 and death was treated as any \r
+other state, i.e. as if it was an observed death at second pass. This was not \r
+precise but correct, although when information on the precise month of\r
+death came (death occuring prior to second pass) we did not change the\r
+likelihood accordingly. We thank Chris Jackson for correcting it. In earlier\r
versions (fortunately before first publication) the total mortality\r
-was overestimated (people were dying too early) of about 10%. Version\r
-0.95 and higher are correct.\r
-\r
-<p> Our suggested choice is mle=1 . If stepm=1 there is no difference\r
-between various mle options (methods of interpolation). If stepm is\r
-big, like 12 or 24 or 48 and mle=4 (no interpolation) the bias may be\r
-very important if the mean duration between two waves is not a\r
-multiple of stepm. See the appendix in our main publication concerning\r
-the sine curve of biases.\r
- \r
-\r
-<h4><font color="#FF0000">Guess values for computing variances</font></h4>\r
-\r
-<p>These values are output by the maximisation of the likelihood <a\r
-href="#mle">mle</a>=1. These valuse can be used as an input of a\r
-second run in order to get the various output data files (Health\r
-expectancies, period prevalence etc.) and figures without rerunning\r
-the long maximisation phase (mle=0). </p>\r
-\r
-<p>These 'scales' are small values needed for the computing of\r
-numerical derivatives. These derivatives are used to compute the\r
-hessian matrix of the parameters, that is the inverse of the\r
-covariance matrix. They are often used for estimating variances and\r
-confidence intervals. Each line consists in indices "ij"\r
-followed by the initial scales (zero to simplify) associated with aij\r
-and bij. </p>\r
-\r
-<ul>\r
- <li>If mle=1 you can enter zeros:</li>\r
- <li><blockquote>\r
- <pre># Scales (for hessian or gradient estimation)\r
+was thus overestimated (people were dying too early) by about 10%. Version\r
+0.95 and higher are correct. \r
+\r
+<P>Our suggested choice is mle=1 . If stepm=1 there is no difference between \r
+various mle options (methods of interpolation). If stepm is big, like 12 or 24 \r
+or 48 and mle=4 (no interpolation) the bias may be very important if the mean \r
+duration between two waves is not a multiple of stepm. See the appendix in our \r
+main publication concerning the sine curve of biases. \r
+<H4><FONT color=#ff0000>Guess values for computing variances</FONT></H4>\r
+<P>These values are output by the maximisation of the likelihood <A \r
+href="http://euroreves.ined.fr/imach/doc/imach.htm#mle">mle</A>=1 and\r
+can be used as an input for a second run in order to get the various output data \r
+files (Health expectancies, period prevalence etc.) and figures without \r
+rerunning the long maximisation phase (mle=0). </P>\r
+<P>The 'scales' are small values needed for the computing of numerical \r
+derivatives. These derivatives are used to compute the hessian matrix of the \r
+parameters, that is the inverse of the covariance matrix. They are often used \r
+for estimating variances and confidence intervals. Each line consists of indices \r
+"ij" followed by the initial scales (zero to simplify) associated with aij and \r
+bij. </P>\r
+<UL>\r
+ <LI>If mle=1 you can enter zeros: \r
+ <LI>\r
+ <BLOCKQUOTE><PRE># Scales (for hessian or gradient estimation)\r
12 0. 0. \r
13 0. 0. \r
21 0. 0. \r
-23 0. 0. </pre>\r
- </blockquote>\r
- </li>\r
- <li>If mle=0 (no maximisation of Likelihood) you must enter a covariance matrix (usually\r
- obtained from an earlier run).</li>\r
-</ul>\r
-\r
-<h4><font color="#FF0000">Covariance matrix of parameters</font></h4>\r
-\r
-<p>The covariance matrix is output if <a href="#mle">mle</a>=1. But it can be\r
-also used as an input to get the various output data files (Health\r
-expectancies, period prevalence etc.) and figures without\r
-rerunning the maximisation phase (mle=0). <br>\r
-Each line starts with indices "ijk" followed by the\r
-covariances between aij and bij:<br>\r
-</p>\r
-\r
-<pre>\r
- 121 Var(a12) \r
+23 0. 0. </PRE></BLOCKQUOTE>\r
+ <LI>If mle=0 (no maximisation of Likelihood) you must enter a covariance \r
+ matrix (usually obtained from an earlier run). </LI></UL>\r
+<H4><FONT color=#ff0000>Covariance matrix of parameters</FONT></H4>\r
+<P>The covariance matrix is output if <A \r
+href="http://euroreves.ined.fr/imach/doc/imach.htm#mle">mle</A>=1. But it can be \r
+also be used as an input to get the various output data files (Health\r
+expectancies, period prevalence etc.) and figures without rerunning\r
+the maximisation phase (mle=0). <BR>Each line starts with indices\r
+"ijk" followed by the covariances between aij and bij:<BR>\r
+</P><PRE> 121 Var(a12) \r
122 Cov(b12,a12) Var(b12) \r
...\r
- 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23) </pre>\r
-\r
-<ul>\r
- <li>If mle=1 you can enter zeros. </li>\r
- <li><pre># Covariance matrix\r
+ 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23) </PRE>\r
+<UL>\r
+ <LI>If mle=1 you can enter zeros. \r
+ <LI><PRE># Covariance matrix\r
121 0.\r
122 0. 0.\r
131 0. 0. 0. \r
211 0. 0. 0. 0. 0. \r
212 0. 0. 0. 0. 0. 0. \r
231 0. 0. 0. 0. 0. 0. 0. \r
-232 0. 0. 0. 0. 0. 0. 0. 0.</pre>\r
- </li>\r
- <li>If mle=0 you must enter a covariance matrix (usually\r
- obtained from an earlier run). </li>\r
-</ul>\r
-\r
-<h4><font color="#FF0000">Age range for calculation of stationary\r
-prevalences and health expectancies</font></h4>\r
-\r
-<pre>agemin=70 agemax=100 bage=50 fage=100</pre>\r
-\r
-<p>\r
-Once we obtained the estimated parameters, the program is able\r
-to calculate period prevalence, transitions probabilities\r
-and life expectancies at any age. Choice of age range is useful\r
-for extrapolation. In this example, age of people interviewed varies\r
-from 69 to 102 and the model is estimated using their exact ages. But\r
-if you are interested in the age-specific period prevalence you can\r
-start the simulation at an exact age like 70 and stop at 100. Then the\r
-program will draw at least two curves describing the forecasted\r
-prevalences of two cohorts, one for healthy people at age 70 and the second\r
-for disabled people at the same initial age. And according to the\r
-mixing property (ergodicity) and because of recovery, both prevalences\r
-will tend to be identical at later ages. Thus if you want to compute\r
-the prevalence at age 70, you should enter a lower agemin value.\r
-\r
-<p>\r
-Setting bage=50 (begin age) and fage=100 (final age), let\r
-the program compute life expectancy from age 'bage' to age\r
-'fage'. As we use a model, we can interessingly compute life\r
-expectancy on a wider age range than the age range from the data.\r
-But the model can be rather wrong on much larger intervals.\r
-Program is limited to around 120 for upper age!\r
-</pre>\r
-\r
-<ul>\r
- <li><b>agemin=</b> Minimum age for calculation of the\r
- period prevalence </li>\r
- <li><b>agemax=</b> Maximum age for calculation of the\r
- period prevalence </li>\r
- <li><b>bage=</b> Minimum age for calculation of the health\r
- expectancies </li>\r
- <li><b>fage=</b> Maximum age for calculation of the health\r
- expectancies </li>\r
-</ul>\r
-\r
-<h4><a name="Computing"><font color="#FF0000">Computing</font></a><font\r
-color="#FF0000"> the cross-sectional prevalence</font></h4>\r
-\r
-<pre>begin-prev-date=1/1/1984 end-prev-date=1/6/1988 estepm=1</pre>\r
-\r
-<p>\r
-Statements 'begin-prev-date' and 'end-prev-date' allow to\r
-select the period in which we calculate the observed prevalences\r
-in each state. In this example, the prevalences are calculated on\r
-data survey collected between 1 january 1984 and 1 june 1988. \r
-</p>\r
-\r
-<ul>\r
- <li><strong>begin-prev-date= </strong>Starting date\r
- (day/month/year)</li>\r
- <li><strong>end-prev-date= </strong>Final date\r
- (day/month/year)</li>\r
- <li><strong>estepm= </strong>Unit (in months).We compute the\r
- life expectancy from trapezoids spaced every estepm\r
- months. This is mainly to measure the difference between\r
- two models: for example if stepm=24 months pijx are given\r
- only every 2 years and by summing them we are calculating\r
- an estimate of the Life Expectancy assuming a linear\r
- progression inbetween and thus overestimating or\r
- underestimating according to the curvature of the\r
- survival function. If, for the same date, we estimate the\r
- model with stepm=1 month, we can keep estepm to 24 months\r
- to compare the new estimate of Life expectancy with the\r
- same linear hypothesis. A more precise result, taking\r
- into account a more precise curvature will be obtained if\r
- estepm is as small as stepm.</li>\r
-</ul>\r
-\r
-<h4><font color="#FF0000">Population- or status-based health\r
-expectancies</font></h4>\r
-\r
-<pre>pop_based=0</pre>\r
-\r
-<p>The program computes status-based health expectancies, i.e health\r
-expectancies which depend on the initial health state. If you are\r
-healthy, your healthy life expectancy (e11) is higher than if you were\r
-disabled (e21, with e11 > e21).<br> To compute a healthy life\r
-expectancy 'independent' of the initial status we have to weight e11\r
-and e21 according to the probability to be in each state at initial\r
-age which are corresponding to the proportions of people in each health\r
-state (cross-sectional prevalences).<p> \r
-\r
-We could also compute e12 and e12 and get e.2 by weighting them\r
-according to the observed cross-sectional prevalences at initial age.\r
-<p> In a similar way we could compute the total life expectancy by\r
-summing e.1 and e.2 .\r
-<br>\r
-The main difference between 'population based' and 'implied' or\r
-'period' consists in the weights used. 'Usually', cross-sectional\r
-prevalences of disability are higher than period prevalences\r
-particularly at old ages. This is true if the country is improving its\r
-health system by teaching people how to prevent disability as by\r
-promoting better screening, for example of people needing cataracts\r
-surgeryand for many unknown reasons that this program may help to\r
-discover. Then the proportion of disabled people at age 90 will be\r
-lower than the current observed proportion.\r
-<p>\r
-Thus a better Health Expectancy and even a better Life Expectancy\r
-value is given by forecasting not only the current lower mortality at\r
-all ages but also a lower incidence of disability and higher recovery.\r
-<br> Using the period prevalences as weight instead of the\r
-cross-sectional prevalences we are computing indices which are more\r
-specific to the current situations and therefore more useful to\r
-predict improvements or regressions in the future as to compare\r
-different policies in various countries.\r
-\r
-<ul>\r
- <li><strong>popbased= 0 </strong>Health expectancies are computed\r
- at each age from period prevalences 'expected' at this initial\r
- age.</li> \r
- <li><strong>popbased= 1 </strong>Health expectancies are\r
- computed at each age from cross-sectional 'observed' prevalence at\r
- this initial age. As all the population is not observed at the\r
- same exact date we define a short period were the observed\r
- prevalence can be computed.<br>\r
-\r
- We simply sum all people surveyed within these two exact dates\r
- who belong to a particular age group (single year) at the date of\r
- interview and being in a particular health state. Then it is easy to\r
-get the proportion of people of a particular health status among all\r
-people of the same age group.<br>\r
-\r
-If both dates are spaced and are covering two waves or more, people\r
-being interviewed twice or more are counted twice or more. The program\r
-takes into account the selection of individuals interviewed between\r
-firstpass and lastpass too (we don't know if it can be useful).\r
-</li>\r
-</ul>\r
-\r
-<h4><font color="#FF0000">Prevalence forecasting (Experimental)</font></h4>\r
-\r
-<pre>starting-proj-date=1/1/1989 final-proj-date=1/1/1992 mov_average=0 </pre>\r
-\r
-<p>Prevalence and population projections are only available if\r
-the interpolation unit is a month, i.e. stepm=1 and if there are\r
-no covariate. The programme estimates the prevalence in each\r
-state at a precise date expressed in day/month/year. The\r
-programme computes one forecasted prevalence a year from a\r
-starting date (1 january of 1989 in this example) to a final date\r
-(1 january 1992). The statement mov_average allows to compute\r
-smoothed forecasted prevalences with a five-age moving average\r
-centered at the mid-age of the five-age period. <br>\r
-</p>\r
-\r
-<h4><font color="#FF0000">Population forecasting (Experimental)</font></h4>\r
-\r
-<ul>\r
- <li><strong>starting-proj-date</strong>= starting date\r
- (day/month/year) of forecasting</li>\r
- <li><strong>final-proj-date= </strong>final date\r
- (day/month/year) of forecasting</li>\r
- <li><strong>mov_average</strong>= smoothing with a five-age\r
- moving average centered at the mid-age of the five-age\r
- period. The command<strong> mov_average</strong> takes\r
- value 1 if the prevalences are smoothed and 0 otherwise.</li>\r
-</ul>\r
-\r
-\r
-<ul type="disc">\r
- <li><b>popforecast=\r
- 0 </b>Option for population forecasting. If\r
- popforecast=1, the programme does the forecasting<b>.</b></li>\r
- <li><b>popfile=\r
- </b>name of the population file</li>\r
- <li><b>popfiledate=</b>\r
- date of the population population</li>\r
- <li><b>last-popfiledate</b>=\r
- date of the last population projection </li>\r
-</ul>\r
-\r
-<hr>\r
-\r
-<h2><a name="running"></a><font color="#00006A">Running Imach\r
-with this example</font></h2>\r
-\r
-<p>We assume that you already typed your <a href="biaspar.imach">1st_example\r
-parameter file</a> as explained <a href="#biaspar">above</a>. \r
-\r
-To run the program under Windows you should either:\r
-</p>\r
-\r
-<ul>\r
- <li>click on the imach.exe icon and either:\r
- <ul>\r
- <li>enter the name of the\r
- parameter file which is for example <tt>\r
-C:\home\myname\lsoa\biaspar.imach"</tt></li>\r
- <li>or locate the biaspar.imach icon in your folder such as\r
- <tt>C:\home\myname\lsoa</tt> \r
- and drag it, with your mouse, on the already open imach window. </li>\r
- </ul>\r
-\r
- <li>With version (0.97b) if you ran setup at installation, Windows is\r
- supposed to understand the ".imach" extension and you can\r
- right click the biaspar.imach icon and either edit with wordpad\r
- (better than notepad) the parameter file or execute it with\r
- IMaCh. </li>\r
-</ul>\r
-\r
-<p>The time to converge depends on the step unit that you used (1\r
-month is more precise but more cpu consuming), on the number of cases,\r
-and on the number of variables (covariates).\r
-\r
-<p>\r
-The program outputs many files. Most of them are files which will be\r
-plotted for better understanding.\r
-\r
-</p>\r
-To run under Linux it is mostly the same.\r
-<p>\r
-It is neither more difficult to run it under a MacIntosh.\r
-<hr>\r
-\r
-<h2><a name="output"><font color="#00006A">Output of the program\r
-and graphs</font> </a></h2>\r
-\r
-<p>Once the optimization is finished (once the convergence is\r
-reached), many tables and graphics are produced.<p>\r
-The IMaCh program will create a subdirectory of the same name as your\r
-parameter file (here mypar) where all the tables and figures will be\r
-stored.<br>\r
-\r
-Important files like the log file and the output parameter file (which\r
-contains the estimates of the maximisation) are stored at the main\r
-level not in this subdirectory. File with extension .log and .txt can\r
-be edited with a standard editor like wordpad or notepad or even can be\r
-viewed with a browser like Internet Explorer or Mozilla.\r
-\r
-<p> The main html file is also named with the same name <a\r
-href="biaspar.htm">biaspar.htm</a>. You can click on it by holding\r
-your shift key in order to open it in another window (Windows).\r
-<p>\r
- Our grapher is Gnuplot, it is an interactive plotting program (GPL) which\r
- can also work in batch. A gnuplot reference manual is available <a\r
- href="http://www.gnuplot.info/">here</a>. <br> When the run is\r
- finished, and in order that the window doesn't disappear, the user\r
- should enter a character like <tt>q</tt> for quitting. <br> These\r
- characters are:<br>\r
-</p>\r
-<ul>\r
- <li>'e' for opening the main result html file <a\r
- href="biaspar.htm"><strong>biaspar.htm</strong></a> file to edit\r
- the output files and graphs. </li> \r
- <li>'g' to graph again</li>\r
- <li>'c' to start again the program from the beginning.</li>\r
- <li>'q' for exiting.</li>\r
-</ul>\r
-\r
-The main gnuplot file is named <tt>biaspar.gp</tt> and can be edited (right\r
-click) and run again.\r
-<p>Gnuplot is easy and you can use it to make more complex\r
-graphs. Just click on gnuplot and type plot sin(x) to see how easy it\r
-is.\r
-\r
-\r
-<h5><font size="4"><strong>Results files </strong></font><br>\r
-<br>\r
-<font color="#EC5E5E" size="3"><strong>- </strong></font><a\r
-name="cross-sectional prevalence in each state"><font color="#EC5E5E"\r
-size="3"><strong>cross-sectional prevalence in each state</strong></font></a><font\r
-color="#EC5E5E" size="3"><strong> (and at first pass)</strong></font><b>:\r
-</b><a href="biaspar/prbiaspar.txt"><b>biaspar/prbiaspar.txt</b></a><br>\r
-</h5>\r
-\r
-<p>The first line is the title and displays each field of the\r
-file. First column corresponds to age. Fields 2 and 6 are the\r
-proportion of individuals in states 1 and 2 respectively as\r
-observed at first exam. Others fields are the numbers of\r
-people in states 1, 2 or more. The number of columns increases if\r
-the number of states is higher than 2.<br>\r
-The header of the file is </p>\r
-\r
-<pre># Age Prev(1) N(1) N Age Prev(2) N(2) N\r
+232 0. 0. 0. 0. 0. 0. 0. 0.</PRE>\r
+ <LI>If mle=0 you must enter a covariance matrix (usually obtained from an \r
+ earlier run). </LI></UL>\r
+<H4><FONT color=#ff0000>Age range for calculation of stationary prevalences and \r
+health expectancies</FONT></H4><PRE>agemin=70 agemax=100 bage=50 fage=100</PRE>\r
+<P>Once we obtained the estimated parameters, the program is able to calculate \r
+period prevalence, transitions probabilities and life expectancies at any age. \r
+Choice of the age range is useful for extrapolation. In this example,\r
+the age of people interviewed varies from 69 to 102 and the model is\r
+estimated using their exact ages. But if you are interested in the\r
+age-specific period prevalence you can start the simulation at an\r
+exact age like 70 and stop at 100. Then the program will draw at\r
+least two curves describing the forecasted prevalences of two cohorts,\r
+one for healthy people at age 70 and the second for disabled people at\r
+the same initial age. And according to the mixing property\r
+(ergodicity) and because of recovery, both prevalences will tend to be\r
+identical at later ages. Thus if you want to compute the prevalence at\r
+age 70, you should enter a lower agemin value. \r
+<P>Setting bage=50 (begin age) and fage=100 (final age), let the program compute \r
+life expectancy from age 'bage' to age 'fage'. As we use a model, we can \r
+interessingly compute life expectancy on a wider age range than the age range \r
+from the data. But the model can be rather wrong on much larger intervals. \r
+Program is limited to around 120 for upper age! <PRE></PRE>\r
+<UL>\r
+ <LI><B>agemin=</B> Minimum age for calculation of the period prevalence \r
+ <LI><B>agemax=</B> Maximum age for calculation of the period prevalence \r
+ <LI><B>bage=</B> Minimum age for calculation of the health expectancies \r
+ <LI><B>fage=</B> Maximum age for calculation of the health expectancies \r
+</LI></UL>\r
+<H4><A name=Computing><FONT color=#ff0000>Computing</FONT></A><FONT \r
+color=#ff0000> the cross-sectional prevalence</FONT></H4><PRE>begin-prev-date=1/1/1984 end-prev-date=1/6/1988 estepm=1</PRE>\r
+<P>Statements 'begin-prev-date' and 'end-prev-date' allow the user to\r
+select the period in which the observed prevalences in each state. In\r
+this example, the prevalences are calculated on data survey collected\r
+between 1 January 1984 and 1 June 1988. </P>\r
+<UL>\r
+ <LI><STRONG>begin-prev-date= </STRONG>Starting date (day/month/year) \r
+ <LI><STRONG>end-prev-date= </STRONG>Final date (day/month/year) \r
+ <LI><STRONG>estepm= </STRONG>Unit (in months).We compute the life expectancy \r
+ from trapezoids spaced every estepm months. This is mainly to measure the \r
+ difference between two models: for example if stepm=24 months pijx are given \r
+ only every 2 years and by summing them we are calculating an estimate of the \r
+ Life Expectancy assuming a linear progression inbetween and thus \r
+ overestimating or underestimating according to the curvature of the survival \r
+ function. If, for the same date, we estimate the model with stepm=1 month, we \r
+ can keep estepm to 24 months to compare the new estimate of Life expectancy \r
+ with the same linear hypothesis. A more precise result, taking into account a \r
+ more precise curvature will be obtained if estepm is as small as stepm. \r
+</LI></UL>\r
+<H4><FONT color=#ff0000>Population- or status-based health \r
+expectancies</FONT></H4><PRE>pop_based=0</PRE>\r
+<P>The program computes status-based health expectancies, i.e health \r
+expectancies which depend on the initial health state. If you are healthy, your \r
+healthy life expectancy (e11) is higher than if you were disabled (e21, with e11 \r
+> e21).<BR>To compute a healthy life expectancy 'independent' of the initial \r
+status we have to weight e11 and e21 according to the probability of\r
+being in each state at initial age which correspond to the proportions\r
+of people in each health state (cross-sectional prevalences).\r
+<P>We could also compute e12 and e12 and get e.2 by weighting them according to \r
+the observed cross-sectional prevalences at initial age. \r
+<P>In a similar way we could compute the total life expectancy by summing e.1 \r
+and e.2 . <BR>The main difference between 'population based' and 'implied' or \r
+'period' is in the weights used. 'Usually', cross-sectional prevalences of \r
+disability are higher than period prevalences particularly at old ages. This is \r
+true if the country is improving its health system by teaching people how to \r
+prevent disability by promoting better screening, for example of people \r
+needing cataract surgery. Then the proportion of disabled people at\r
+age 90 will be lower than the current observed proportion. \r
+<P>Thus a better Health Expectancy and even a better Life Expectancy value is \r
+given by forecasting not only the current lower mortality at all ages but also a \r
+lower incidence of disability and higher recovery. <BR>Using the period \r
+prevalences as weight instead of the cross-sectional prevalences we are \r
+computing indices which are more specific to the current situations and \r
+therefore more useful to predict improvements or regressions in the future as to \r
+compare different policies in various countries. \r
+<UL>\r
+ <LI><STRONG>popbased= 0 </STRONG>Health expectancies are computed at each age \r
+ from period prevalences 'expected' at this initial age. \r
+ <LI><STRONG>popbased= 1 </STRONG>Health expectancies are computed at each age \r
+ from cross-sectional 'observed' prevalence at the initial age. As all the \r
+ population is not observed at the same exact date we define a short period \r
+ where the observed prevalence can be computed as follows:<BR>we simply sum all people \r
+ surveyed within these two exact dates who belong to a particular age group \r
+ (single year) at the date of interview and are in a particular health state. \r
+ Then it is easy to get the proportion of people in a particular\r
+ health state as a percentage of all people of the same age group.<BR>If both dates are spaced and are \r
+ covering two waves or more, people being interviewed twice or more are counted \r
+ twice or more. The program takes into account the selection of individuals \r
+ interviewed between firstpass and lastpass too (we don't know if\r
+ this is useful). </LI></UL>\r
+<H4><FONT color=#ff0000>Prevalence forecasting (Experimental)</FONT></H4><PRE>starting-proj-date=1/1/1989 final-proj-date=1/1/1992 mov_average=0 </PRE>\r
+<P>Prevalence and population projections are only available if the interpolation \r
+unit is a month, i.e. stepm=1 and if there are no covariate. The programme \r
+estimates the prevalence in each state at a precise date expressed in \r
+day/month/year. The programme computes one forecasted prevalence a year from a \r
+starting date (1 January 1989 in this example) to a final date (1 January \r
+1992). The statement mov_average allows computation of smoothed forecasted \r
+prevalences with a five-age moving average centered at the mid-age of the \r
+fiveyear-age period. <BR></P>\r
+<H4><FONT color=#ff0000>Population forecasting (Experimental)</FONT></H4>\r
+<UL>\r
+ <LI><STRONG>starting-proj-date</STRONG>= starting date (day/month/year) of \r
+ forecasting \r
+ <LI><STRONG>final-proj-date= </STRONG>final date (day/month/year) of \r
+ forecasting \r
+ <LI><STRONG>mov_average</STRONG>= smoothing with a five-age moving average \r
+ centered at the mid-age of the fiveyear-age period. The command<STRONG> \r
+ mov_average</STRONG> takes value 1 if the prevalences are smoothed and 0 \r
+ otherwise. </LI></UL>\r
+<UL type=disc>\r
+ <LI><B>popforecast= 0 </B>Option for population forecasting. If popforecast=1, \r
+ the programme does the forecasting<B>.</B> \r
+ <LI><B>popfile= </B>name of the population file \r
+ <LI><B>popfiledate=</B> date of the population population \r
+ <LI><B>last-popfiledate</B>= date of the last population projection \r
+</LI></UL>\r
+<HR>\r
+\r
+<H2><A name=running></A><FONT color=#00006a>Running Imach with this \r
+example</FONT></H2>\r
+<P>We assume that you have already typed your <A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar.imach">1st_example parameter \r
+file</A> as explained <A \r
+href="http://euroreves.ined.fr/imach/doc/imach.htm#biaspar">above</A>. To run \r
+the program under Windows you should either: </P>\r
+<UL>\r
+ <LI>click on the imach.exe icon and either: \r
+ <UL>\r
+ <LI>enter the name of the parameter file which is for example \r
+ <TT>C:\home\myname\lsoa\biaspar.imach</TT> \r
+ <LI>or locate the biaspar.imach icon in your folder such as \r
+ <TT>C:\home\myname\lsoa</TT> and drag it, with your mouse, on the already \r
+ open imach window. </LI></UL>\r
+ <LI>With version (0.97b) if you ran setup at installation, Windows is supposed \r
+ to understand the ".imach" extension and you can right click the biaspar.imach \r
+ icon and either edit with wordpad (better than notepad) the parameter file or \r
+ execute it with IMaCh. </LI></UL>\r
+<P>The time to converge depends on the step unit used (1 month is more \r
+precise but more cpu time consuming), on the number of cases, and on the number of \r
+variables (covariates). \r
+<P>The program outputs many files. Most of them are files which will be plotted \r
+for better understanding. </P>To run under Linux is mostly the same. \r
+<P>It is no more difficult to run IMaCh on a MacIntosh. \r
+<HR>\r
+\r
+<H2><A name=output><FONT color=#00006a>Output of the program and graphs</FONT> \r
+</A></H2>\r
+<P>Once the optimization is finished (once the convergence is reached), many \r
+tables and graphics are produced.\r
+<P>The IMaCh program will create a subdirectory with the same name as your \r
+parameter file (here mypar) where all the tables and figures will be \r
+stored.<BR>Important files like the log file and the output parameter file \r
+(the latter contains the maximum likelihood estimates) are stored at\r
+the main level not in this subdirectory. Files with extension .log and\r
+.txt can be edited with a standard editor like wordpad or notepad or\r
+even can be viewed with a browser like Internet Explorer or Mozilla. \r
+<P>The main html file is also named with the same name <A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar.htm">biaspar.htm</A>. You can \r
+click on it by holding your shift key in order to open it in another window \r
+(Windows). \r
+<P>Our grapher is Gnuplot, an interactive plotting program (GPL) which can \r
+also work in batch mode. A gnuplot reference manual is available <A \r
+href="http://www.gnuplot.info/">here</A>. <BR>When the run is finished, and in \r
+order that the window doesn't disappear, the user should enter a character like \r
+<TT>q</TT> for quitting. <BR>These characters are:<BR></P>\r
+<UL>\r
+ <LI>'e' for opening the main result html file <A \r
+ href="http://euroreves.ined.fr/imach/doc/biaspar.htm"><STRONG>biaspar.htm</STRONG></A> \r
+ file to edit the output files and graphs. \r
+ <LI>'g' to graph again \r
+ <LI>'c' to start again the program from the beginning. \r
+ <LI>'q' for exiting. </LI></UL>The main gnuplot file is named \r
+<TT>biaspar.gp</TT> and can be edited (right click) and run again. \r
+<P>Gnuplot is easy and you can use it to make more complex graphs. Just click on \r
+gnuplot and type plot sin(x) to see how easy it is. \r
+<H5><FONT size=4><STRONG>Results files </STRONG></FONT><BR><BR><FONT \r
+color=#ec5e5e size=3><STRONG>- </STRONG></FONT><A \r
+name="cross-sectional prevalence in each state"><FONT color=#ec5e5e \r
+size=3><STRONG>cross-sectional prevalence in each state</STRONG></FONT></A><FONT \r
+color=#ec5e5e size=3><STRONG> (and at first pass)</STRONG></FONT><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/prbiaspar.txt"><B>biaspar/prbiaspar.txt</B></A><BR></H5>\r
+<P>The first line is the title and displays each field of the file. First column \r
+corresponds to age. Fields 2 and 6 are the proportion of individuals in states 1 \r
+and 2 respectively as observed at first exam. Others fields are the numbers of \r
+people in states 1, 2 or more. The number of columns increases if the number of \r
+states is higher than 2.<BR>The header of the file is </P><PRE># Age Prev(1) N(1) N Age Prev(2) N(2) N\r
70 1.00000 631 631 70 0.00000 0 631\r
71 0.99681 625 627 71 0.00319 2 627 \r
-72 0.97125 1115 1148 72 0.02875 33 1148 </pre>\r
-\r
-<p>It means that at age 70 (between 70 and 71), the prevalence in state 1 is 1.000\r
-and in state 2 is 0.00 . At age 71 the number of individuals in\r
-state 1 is 625 and in state 2 is 2, hence the total number of\r
-people aged 71 is 625+2=627. <br>\r
-</p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- Estimated parameters and\r
-covariance matrix</b></font><b>: </b><a href="rbiaspar.txt"><b>rbiaspar.imach</b></a></h5>\r
-\r
-<p>This file contains all the maximisation results: </p>\r
-\r
-<pre> -2 log likelihood= 21660.918613445392\r
+72 0.97125 1115 1148 72 0.02875 33 1148 </PRE>\r
+<P>It means that at age 70 (between 70 and 71), the prevalence in state 1 is \r
+1.000 and in state 2 is 0.00 . At age 71 the number of individuals in state 1 is \r
+625 and in state 2 is 2, hence the total number of people aged 71 is 625+2=627. \r
+<BR></P>\r
+<H5><FONT color=#ec5e5e size=3><B>- Estimated parameters and covariance \r
+matrix</B></FONT><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/rbiaspar.txt"><B>rbiaspar.imach</B></A></H5>\r
+<P>This file contains all the maximisation results: </P><PRE> -2 log likelihood= 21660.918613445392\r
Estimated parameters: a12 = -12.290174 b12 = 0.092161 \r
a13 = -9.155590 b13 = 0.046627 \r
a21 = -2.629849 b21 = -0.022030 \r
Var(b21) = 1.29229e-004\r
Var(a23) = 4.48405e-001\r
Var(b23) = 5.85631e-005 \r
- </pre>\r
-\r
-<p>By substitution of these parameters in the regression model,\r
-we obtain the elementary transition probabilities:</p>\r
-\r
-<p><img src="biaspar/pebiaspar11.png" width="400" height="300"></p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- Transition probabilities</b></font><b>:\r
-</b><a href="biaspar/pijrbiaspar.txt"><b>biaspar/pijrbiaspar.txt</b></a></h5>\r
-\r
-<p>Here are the transitions probabilities Pij(x, x+nh). The second\r
-column is the starting age x (from age 95 to 65), the third is age\r
-(x+nh) and the others are the transition probabilities p11, p12, p13,\r
-p21, p22, p23. The first column indicates the value of the covariate\r
-(without any other variable than age it is equal to 1) For example, line 5 of the file\r
-is: </p>\r
-\r
-<pre>1 100 106 0.02655 0.17622 0.79722 0.01809 0.13678 0.84513 </pre>\r
-\r
-<p>and this means: </p>\r
-\r
-<pre>p11(100,106)=0.02655\r
+ </PRE>\r
+<P>By substitution of these parameters in the regression model, we obtain the \r
+elementary transition probabilities:</P>\r
+<P><IMG height=300 \r
+src="Computing Health Expectancies using IMaCh_fichiers/pebiaspar11.png" \r
+width=400></P>\r
+<H5><FONT color=#ec5e5e size=3><B>- Transition probabilities</B></FONT><B>: \r
+</B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/pijrbiaspar.txt"><B>biaspar/pijrbiaspar.txt</B></A></H5>\r
+<P>Here are the transitions probabilities Pij(x, x+nh). The second column is the \r
+starting age x (from age 95 to 65), the third is age (x+nh) and the others are \r
+the transition probabilities p11, p12, p13, p21, p22, p23. The first column \r
+indicates the value of the covariate (without any other variable than age it is \r
+equal to 1) For example, line 5 of the file is: </P><PRE>1 100 106 0.02655 0.17622 0.79722 0.01809 0.13678 0.84513 </PRE>\r
+<P>and this means: </P><PRE>p11(100,106)=0.02655\r
p12(100,106)=0.17622\r
p13(100,106)=0.79722\r
p21(100,106)=0.01809\r
p22(100,106)=0.13678\r
-p22(100,106)=0.84513 </pre>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- </b></font><a\r
-name="Period prevalence in each state"><font color="#EC5E5E"\r
-size="3"><b>Period prevalence in each state</b></font></a><b>:\r
-</b><a href="biaspar/plrbiaspar.txt"><b>biaspar/plrbiaspar.txt</b></a></h5>\r
-\r
-<pre>#Prevalence\r
+p22(100,106)=0.84513 </PRE>\r
+<H5><FONT color=#ec5e5e size=3><B>- </B></FONT><A \r
+name="Period prevalence in each state"><FONT color=#ec5e5e size=3><B>Period \r
+prevalence in each state</B></FONT></A><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/plrbiaspar.txt"><B>biaspar/plrbiaspar.txt</B></A></H5><PRE>#Prevalence\r
#Age 1-1 2-2\r
\r
#************ \r
70 0.90134 0.09866\r
71 0.89177 0.10823 \r
72 0.88139 0.11861 \r
-73 0.87015 0.12985 </pre>\r
-\r
-<p>At age 70 the period prevalence is 0.90134 in state 1 and 0.09866\r
-in state 2. This period prevalence differs from the cross-sectional\r
-prevalence. Here is the point. The cross-sectional prevalence at age\r
-70 results from the incidence of disability, incidence of recovery and\r
-mortality which occurred in the past of the cohort. Period prevalence\r
-results from a simulation with current incidences of disability,\r
-recovery and mortality estimated from this cross-longitudinal\r
-survey. It is a good predictin of the prevalence in the\r
-future if "nothing changes in the future". This is exactly\r
-what demographers do with a period life table. Life expectancy is the\r
-expected mean survival time if current mortality rates (age-specific incidences\r
-of mortality) "remain constant" in the future. </p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- Standard deviation of\r
-period prevalence</b></font><b>: </b><a\r
-href="biaspar/vplrbiaspar.txt"><b>biaspar/vplrbiaspar.txt</b></a></h5>\r
-\r
-<p>The period prevalence has to be compared with the cross-sectional\r
-prevalence. But both are statistical estimates and therefore\r
-have confidence intervals.\r
-<br>For the cross-sectional prevalence we generally need information on\r
-the design of the surveys. It is usually not enough to consider the\r
-number of people surveyed at a particular age and to estimate a\r
-Bernouilli confidence interval based on the prevalence at that\r
-age. But you can do it to have an idea of the randomness. At least you\r
-can get a visual appreciation of the randomness by looking at the\r
-fluctuation over ages.\r
-\r
-<p> For the period prevalence it is possible to estimate the\r
-confidence interval from the Hessian matrix (see the publication for\r
-details). We are supposing that the design of the survey will only\r
-alter the weight of each individual. IMaCh is scaling the weights of\r
-individuals-waves contributing to the likelihood by making the sum of\r
-the weights equal to the sum of individuals-waves contributing: a\r
-weighted survey doesn't increase or decrease the size of the survey,\r
-it only give more weights to some individuals and thus less to the\r
-others.\r
-\r
-<h5><font color="#EC5E5E" size="3">-cross-sectional and period\r
-prevalence in state (2=disable) with confidence interval</font>:<b>\r
-</b><a href="biaspar/vbiaspar21.htm"><b>biaspar/vbiaspar21.png</b></a></h5>\r
-\r
-<p>This graph exhibits the period prevalence in state (2) with the\r
-confidence interval in red. The green curve is the observed prevalence\r
-(or proportion of individuals in state (2)). Without discussing the\r
-results (it is not the purpose here), we observe that the green curve\r
-is rather below the period prevalence. It the data where not biased by\r
-the non inclusion of people living in institutions we would have\r
-concluded that the prevalence of disability will increase in the\r
-future (see the main publication if you are interested in real data\r
-and results which are opposite).</p>\r
-\r
-<p><img src="biaspar/vbiaspar21.png" width="400" height="300"></p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>-Convergence to the\r
-period prevalence of disability</b></font><b>: </b><a\r
-href="biaspar/pbiaspar11.png"><b>biaspar/pbiaspar11.png</b></a><br>\r
-<img src="biaspar/pbiaspar11.png" width="400" height="300"> </h5>\r
-\r
-<p>This graph plots the conditional transition probabilities from\r
-an initial state (1=healthy in red at the bottom, or 2=disable in\r
-green on top) at age <em>x </em>to the final state 2=disable<em> </em>at\r
-age <em>x+h. </em>Conditional means at the condition to be alive\r
-at age <em>x+h </em>which is <i>hP12x</i> + <em>hP22x</em>. The\r
-curves <i>hP12x/(hP12x</i> + <em>hP22x) </em>and <i>hP22x/(hP12x</i>\r
-+ <em>hP22x) </em>converge with <em>h, </em>to the <em>period\r
-prevalence of disability</em>. In order to get the period\r
-prevalence at age 70 we should start the process at an earlier\r
-age, i.e.50. If the disability state is defined by severe\r
-disability criteria with only a few chance to recover, then the\r
-incidence of recovery is low and the time to convergence is\r
-probably longer. But we don't have experience yet.</p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- Life expectancies by age\r
-and initial health status with standard deviation</b></font><b>: </b><a\r
-href="biaspar/erbiaspar.txt"><b>biaspar/erbiaspar.txt</b></a></h5>\r
-\r
-<pre># Health expectancies \r
+73 0.87015 0.12985 </PRE>\r
+<P>At age 70 the period prevalence is 0.90134 in state 1 and 0.09866 in state 2. \r
+This period prevalence differs from the cross-sectional prevalence and\r
+we explaining. The cross-sectional prevalence at age 70 results from\r
+the incidence of disability, incidence of recovery and mortality which\r
+occurred in the past for the cohort. Period prevalence results from a\r
+simulation with current incidences of disability, recovery and\r
+mortality estimated from this cross-longitudinal survey. It is a good\r
+prediction of the prevalence in the future if "nothing changes in the\r
+future". This is exactly what demographers do with a period life\r
+table. Life expectancy is the expected mean survival time if current\r
+mortality rates (age-specific incidences of mortality) "remain\r
+constant" in the future. \r
+</P>\r
+<H5><FONT color=#ec5e5e size=3><B>- Standard deviation of period \r
+prevalence</B></FONT><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/vplrbiaspar.txt"><B>biaspar/vplrbiaspar.txt</B></A></H5>\r
+<P>The period prevalence has to be compared with the cross-sectional prevalence. \r
+But both are statistical estimates and therefore have confidence intervals. \r
+<BR>For the cross-sectional prevalence we generally need information on the \r
+design of the surveys. It is usually not enough to consider the number of people \r
+surveyed at a particular age and to estimate a Bernouilli confidence interval \r
+based on the prevalence at that age. But you can do it to have an idea of the \r
+randomness. At least you can get a visual appreciation of the randomness by \r
+looking at the fluctuation over ages. \r
+<P>For the period prevalence it is possible to estimate the confidence interval \r
+from the Hessian matrix (see the publication for details). We are supposing that \r
+the design of the survey will only alter the weight of each individual. IMaCh \r
+scales the weights of individuals-waves contributing to the likelihood by \r
+making the sum of the weights equal to the sum of individuals-waves \r
+contributing: a weighted survey doesn't increase or decrease the size of the \r
+survey, it only give more weight to some individuals and thus less to the \r
+others. \r
+<H5><FONT color=#ec5e5e size=3>-cross-sectional and period prevalence in state \r
+(2=disable) with confidence interval</FONT>:<B> </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/vbiaspar21.htm"><B>biaspar/vbiaspar21.png</B></A></H5>\r
+<P>This graph exhibits the period prevalence in state (2) with the confidence \r
+interval in red. The green curve is the observed prevalence (or proportion of \r
+individuals in state (2)). Without discussing the results (it is not the purpose \r
+here), we observe that the green curve is somewhat below the period\r
+prevalence. If the data were not biased by the non inclusion of people\r
+living in institutions we would have concluded that the prevalence of\r
+disability will increase in the future (see the main publication if\r
+you are interested in real data and results which are opposite).</P>\r
+<P><IMG height=300 \r
+src="Computing Health Expectancies using IMaCh_fichiers/vbiaspar21.png" \r
+width=400></P>\r
+<H5><FONT color=#ec5e5e size=3><B>-Convergence to the period prevalence of \r
+disability</B></FONT><B>: </B><A \r
+href="Computing Health Expectancies using IMaCh_fichiers/pbiaspar11.png"><B>biaspar/pbiaspar11.png</B></A><BR><IMG \r
+height=300 \r
+src="Computing Health Expectancies using IMaCh_fichiers/pbiaspar11.png" \r
+width=400> </H5>\r
+<P>This graph plots the conditional transition probabilities from an initial \r
+state (1=healthy in red at the bottom, or 2=disabled in green on the top) at age \r
+<EM>x </EM>to the final state 2=disabled<EM> </EM>at age <EM>x+h\r
+</EM> where conditional means conditional on being alive at age <EM>x+h </EM>which is \r
+<I>hP12x</I> + <EM>hP22x</EM>. The curves <I>hP12x/(hP12x</I> + <EM>hP22x) \r
+</EM>and <I>hP22x/(hP12x</I> + <EM>hP22x) </EM>converge with <EM>h, </EM>to the \r
+<EM>period prevalence of disability</EM>. In order to get the period prevalence \r
+at age 70 we should start the process at an earlier age, i.e.50. If the \r
+disability state is defined by severe disability criteria with only a\r
+small chance of recovering, then the incidence of recovery is low and the time to convergence is \r
+probably longer. But we don't have experience of this yet.</P>\r
+<H5><FONT color=#ec5e5e size=3><B>- Life expectancies by age and initial health \r
+status with standard deviation</B></FONT><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/erbiaspar.txt"><B>biaspar/erbiaspar.txt</B></A></H5><PRE># Health expectancies \r
# Age 1-1 (SE) 1-2 (SE) 2-1 (SE) 2-2 (SE)\r
70 11.0180 (0.1277) 3.1950 (0.3635) 4.6500 (0.0871) 4.4807 (0.2187)\r
71 10.4786 (0.1184) 3.2093 (0.3212) 4.3384 (0.0875) 4.4820 (0.2076)\r
79 6.7464 (0.0867) 3.3220 (0.1124) 2.3794 (0.1112) 4.4646 (0.1364)\r
80 6.3538 (0.0868) 3.3354 (0.1014) 2.1949 (0.1168) 4.4587 (0.1331)\r
81 5.9775 (0.0873) 3.3484 (0.0933) 2.0222 (0.1230) 4.4520 (0.1320)\r
-</pre>\r
-\r
-<pre>For example 70 11.0180 (0.1277) 3.1950 (0.3635) 4.6500 (0.0871) 4.4807 (0.2187)\r
+</PRE><PRE>For example 70 11.0180 (0.1277) 3.1950 (0.3635) 4.6500 (0.0871) 4.4807 (0.2187)\r
means\r
-e11=11.0180 e12=3.1950 e21=4.6500 e22=4.4807 </pre>\r
-\r
-<pre><img src="biaspar/expbiaspar21.png" width="400" height="300"><img\r
-src="biaspar/expbiaspar11.png" width="400" height="300"></pre>\r
-\r
-<p>For example, life expectancy of a healthy individual at age 70\r
-is 11.0 in the healthy state and 3.2 in the disability state\r
-(total of 14.2 years). If he was disable at age 70, his life expectancy\r
-will be shorter, 4.65 years in the healthy state and 4.5 in the\r
-disability state (=9.15 years). The total life expectancy is a\r
-weighted mean of both, 14.2 and 9.15. The weight is the proportion\r
-of people disabled at age 70. In order to get a period index\r
-(i.e. based only on incidences) we use the <a\r
-href="#Period prevalence in each state">stable or\r
-period prevalence</a> at age 70 (i.e. computed from\r
-incidences at earlier ages) instead of the <a\r
-href="#cross-sectional prevalence in each state">cross-sectional prevalence</a>\r
-(observed for example at first medical exam) (<a href="#Health expectancies">see\r
-below</a>).</p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- Variances of life\r
-expectancies by age and initial health status</b></font><b>: </b><a\r
-href="biaspar/vrbiaspar.txt"><b>biaspar/vrbiaspar.txt</b></a></h5>\r
-\r
-<p>For example, the covariances of life expectancies Cov(ei,ej)\r
-at age 50 are (line 3) </p>\r
-\r
-<pre> Cov(e1,e1)=0.4776 Cov(e1,e2)=0.0488=Cov(e2,e1) Cov(e2,e2)=0.0424</pre>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>-Variances of one-step\r
-probabilities </b></font><b>: </b><a href="biaspar/probrbiaspar.txt"><b>biaspar/probrbiaspar.txt</b></a></h5>\r
-\r
-<p>For example, at age 65</p>\r
-\r
-<pre> p11=9.960e-001 standard deviation of p11=2.359e-004</pre>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- </b></font><a\r
-name="Health expectancies"><font color="#EC5E5E" size="3"><b>Health\r
-expectancies</b></font></a><font color="#EC5E5E" size="3"><b>\r
-with standard errors in parentheses</b></font><b>: </b><a\r
-href="biaspar/trbiaspar.txt"><font face="Courier New"><b>biaspar/trbiaspar.txt</b></font></a></h5>\r
-\r
-<pre>#Total LEs with variances: e.. (std) e.1 (std) e.2 (std) </pre>\r
-\r
-<pre>70 13.26 (0.22) 9.95 (0.20) 3.30 (0.14) </pre>\r
-\r
-<p>Thus, at age 70 the total life expectancy, e..=13.26 years is\r
-the weighted mean of e1.=13.46 and e2.=11.35 by the period\r
-prevalences at age 70 which are 0.90134 in state 1 and 0.09866 in\r
-state 2 respectively (the sum is equal to one). e.1=9.95 is the\r
-Disability-free life expectancy at age 70 (it is again a weighted\r
-mean of e11 and e21). e.2=3.30 is also the life expectancy at age\r
-70 to be spent in the disability state.</p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>-Total life expectancy by\r
-age and health expectancies in states (1=healthy) and (2=disable)</b></font><b>:\r
-</b><a href="biaspar/ebiaspar1.png"><b>biaspar/ebiaspar1.png</b></a></h5>\r
-\r
-<p>This figure represents the health expectancies and the total\r
-life expectancy with a confidence interval (dashed line). </p>\r
-\r
-<pre> <img src="biaspar/ebiaspar1.png" width="400" height="300"></pre>\r
-\r
-<p>Standard deviations (obtained from the information matrix of\r
-the model) of these quantities are very useful.\r
-Cross-longitudinal surveys are costly and do not involve huge\r
-samples, generally a few thousands; therefore it is very\r
-important to have an idea of the standard deviation of our\r
-estimates. It has been a big challenge to compute the Health\r
-Expectancy standard deviations. Don't be confuse: life expectancy\r
-is, as any expected value, the mean of a distribution; but here\r
-we are not computing the standard deviation of the distribution,\r
-but the standard deviation of the estimate of the mean.</p>\r
-\r
-<p>Our health expectancies estimates vary according to the sample\r
-size (and the standard deviations give confidence intervals of\r
-the estimates) but also according to the model fitted. Let us\r
-explain it in more details.</p>\r
-\r
-<p>Choosing a model means at least two kind of choices. At first we\r
-have to decide the number of disability states. And at second we have to\r
-design, within the logit model family, the model itself: variables,\r
-covariables, confounding factors etc. to be included.</p>\r
-\r
-<p>More disability states we have, better is our demographical\r
-approach of the disability process, but smaller are the number of\r
-transitions between each state and higher is the noise in the\r
-measurement. We do not have enough experiments of the various\r
-models to summarize the advantages and disadvantages, but it is\r
-important to say that even if we had huge and unbiased samples,\r
-the total life expectancy computed from a cross-longitudinal\r
-survey, varies with the number of states. If we define only two\r
-states, alive or dead, we find the usual life expectancy where it\r
-is assumed that at each age, people are at the same risk to die.\r
-If we are differentiating the alive state into healthy and\r
-disable, and as the mortality from the disability state is higher\r
-than the mortality from the healthy state, we are introducing\r
-heterogeneity in the risk of dying. The total mortality at each\r
-age is the weighted mean of the mortality in each state by the\r
-prevalence in each state. Therefore if the proportion of people\r
-at each age and in each state is different from the period\r
-equilibrium, there is no reason to find the same total mortality\r
-at a particular age. Life expectancy, even if it is a very useful\r
-tool, has a very strong hypothesis of homogeneity of the\r
-population. Our main purpose is not to measure differential\r
-mortality but to measure the expected time in a healthy or\r
-disability state in order to maximise the former and minimize the\r
-latter. But the differential in mortality complexifies the\r
-measurement.</p>\r
-\r
-<p>Incidences of disability or recovery are not affected by the number\r
-of states if these states are independent. But incidences estimates\r
-are dependent on the specification of the model. More covariates we\r
-added in the logit model better is the model, but some covariates are\r
-not well measured, some are confounding factors like in any\r
-statistical model. The procedure to "fit the best model' is\r
-similar to logistic regression which itself is similar to regression\r
-analysis. We haven't yet been sofar because we also have a severe\r
-limitation which is the speed of the convergence. On a Pentium III,\r
-500 MHz, even the simplest model, estimated by month on 8,000 people\r
-may take 4 hours to converge. Also, the IMaCh program is not a\r
-statistical package, and does not allow sophisticated design\r
-variables. If you need sophisticated design variable you have to them\r
-your self and and add them as ordinary variables. IMaCX allows up to 8\r
-variables. The current version of this program allows only to add\r
-simple variables like age+sex or age+sex+ age*sex but will never be\r
-general enough. But what is to remember, is that incidences or\r
-probability of change from one state to another is affected by the\r
-variables specified into the model.</p>\r
-\r
-<p>Also, the age range of the people interviewed is linked \r
-the age range of the life expectancy which can be estimated by\r
-extrapolation. If your sample ranges from age 70 to 95, you can\r
-clearly estimate a life expectancy at age 70 and trust your\r
-confidence interval because it is mostly based on your sample size,\r
-but if you want to estimate the life expectancy at age 50, you\r
-should rely in the design of your model. Fitting a logistic model on a age\r
-range of 70 to 95 and estimating probabilties of transition out of\r
-this age range, say at age 50, is very dangerous. At least you\r
-should remember that the confidence interval given by the\r
-standard deviation of the health expectancies, are under the\r
-strong assumption that your model is the 'true model', which is\r
-probably not the case outside the age range of your sample.</p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- Copy of the parameter\r
-file</b></font><b>: </b><a href="orbiaspar.txt"><b>orbiaspar.txt</b></a></h5>\r
-\r
-<p>This copy of the parameter file can be useful to re-run the\r
-program while saving the old output files. </p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- Prevalence forecasting</b></font><b>:\r
-</b><a href="biaspar/frbiaspar.txt"><b>biaspar/frbiaspar.txt</b></a></h5>\r
-\r
-<p>\r
-\r
-First,\r
-we have estimated the observed prevalence between 1/1/1984 and\r
-1/6/1988 (June, European syntax of dates). The mean date of all interviews (weighted average of the\r
-interviews performed between 1/1/1984 and 1/6/1988) is estimated\r
-to be 13/9/1985, as written on the top on the file. Then we\r
-forecast the probability to be in each state. </p>\r
-\r
-<p>\r
-For example on 1/1/1989 : </p>\r
-\r
-<pre class="MsoNormal"># StartingAge FinalAge P.1 P.2 P.3\r
+e11=11.0180 e12=3.1950 e21=4.6500 e22=4.4807 </PRE><PRE><IMG height=300 src="Computing Health Expectancies using IMaCh_fichiers/expbiaspar21.png" width=400><IMG height=300 src="Computing Health Expectancies using IMaCh_fichiers/expbiaspar11.png" width=400></PRE>\r
+<P>For example, life expectancy of a healthy individual at age 70 is 11.0 in the \r
+healthy state and 3.2 in the disability state (total of 14.2 years). If he was \r
+disabled at age 70, his life expectancy will be shorter, 4.65 years in the \r
+healthy state and 4.5 in the disability state (=9.15 years). The total life \r
+expectancy is a weighted mean of both, 14.2 and 9.15. The weight is the \r
+proportion of people disabled at age 70. In order to get a period index (i.e. \r
+based only on incidences) we use the <A \r
+href="http://euroreves.ined.fr/imach/doc/imach.htm#Period prevalence in each state">stable \r
+or period prevalence</A> at age 70 (i.e. computed from incidences at earlier \r
+ages) instead of the <A \r
+href="http://euroreves.ined.fr/imach/doc/imach.htm#cross-sectional prevalence in each state">cross-sectional \r
+prevalence</A> (observed for example at first interview) (<A \r
+href="http://euroreves.ined.fr/imach/doc/imach.htm#Health expectancies">see \r
+below</A>).</P>\r
+<H5><FONT color=#ec5e5e size=3><B>- Variances of life expectancies by age and \r
+initial health status</B></FONT><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/vrbiaspar.txt"><B>biaspar/vrbiaspar.txt</B></A></H5>\r
+<P>For example, the covariances of life expectancies Cov(ei,ej) at age 50 are \r
+(line 3) </P><PRE> Cov(e1,e1)=0.4776 Cov(e1,e2)=0.0488=Cov(e2,e1) Cov(e2,e2)=0.0424</PRE>\r
+<H5><FONT color=#ec5e5e size=3><B>-Variances of one-step probabilities \r
+</B></FONT><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/probrbiaspar.txt"><B>biaspar/probrbiaspar.txt</B></A></H5>\r
+<P>For example, at age 65</P><PRE> p11=9.960e-001 standard deviation of p11=2.359e-004</PRE>\r
+<H5><FONT color=#ec5e5e size=3><B>- </B></FONT><A \r
+name="Health expectancies"><FONT color=#ec5e5e size=3><B>Health \r
+expectancies</B></FONT></A><FONT color=#ec5e5e size=3><B> with standard errors \r
+in parentheses</B></FONT><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/trbiaspar.txt"><FONT \r
+face="Courier New"><B>biaspar/trbiaspar.txt</B></FONT></A></H5><PRE>#Total LEs with variances: e.. (std) e.1 (std) e.2 (std) </PRE><PRE>70 13.26 (0.22) 9.95 (0.20) 3.30 (0.14) </PRE>\r
+<P>Thus, at age 70 the total life expectancy, e..=13.26 years is the weighted \r
+mean of e1.=13.46 and e2.=11.35 by the period prevalences at age 70 which are \r
+0.90134 in state 1 and 0.09866 in state 2 respectively (the sum is equal to \r
+one). e.1=9.95 is the Disability-free life expectancy at age 70 (it is again a \r
+weighted mean of e11 and e21). e.2=3.30 is also the life expectancy at age 70 to \r
+be spent in the disability state.</P>\r
+<H5><FONT color=#ec5e5e size=3><B>-Total life expectancy by age and health \r
+expectancies in states (1=healthy) and (2=disable)</B></FONT><B>: </B><A \r
+href="Computing Health Expectancies using IMaCh_fichiers/ebiaspar1.png"><B>biaspar/ebiaspar1.png</B></A></H5>\r
+<P>This figure represents the health expectancies and the total life expectancy \r
+with a confidence interval (dashed line). </P><PRE> <IMG height=300 src="Computing Health Expectancies using IMaCh_fichiers/ebiaspar1.png" width=400></PRE>\r
+<P>Standard deviations (obtained from the information matrix of the model) of \r
+these quantities are very useful. Cross-longitudinal surveys are costly and do \r
+not involve huge samples, generally a few thousands; therefore it is very \r
+important to have an idea of the standard deviation of our estimates. It has \r
+been a big challenge to compute the Health Expectancy standard deviations. Don't \r
+be confused: life expectancy is, as any expected value, the mean of a \r
+distribution; but here we are not computing the standard deviation of the \r
+distribution, but the standard deviation of the estimate of the mean.</P>\r
+<P>Our health expectancy estimates vary according to the sample size (and the \r
+standard deviations give confidence intervals of the estimates) but also \r
+according to the model fitted. We explain this in more detail.</P>\r
+<P>Choosing a model means at least two kind of choices. First we have to \r
+decide the number of disability states. And second we have to design, within \r
+the logit model family, the model itself: variables, covariates, confounding \r
+factors etc. to be included.</P>\r
+<P>The more disability states we have, the better is our demographical\r
+approximation of the disability process, but the smaller the number of\r
+transitions between each state and the higher the noise in the\r
+measurement. We have not experimented enough with the various models\r
+to summarize the advantages and disadvantages, but it is important to\r
+note that even if we had huge unbiased samples, the total life\r
+expectancy computed from a cross-longitudinal survey would vary with\r
+the number of states. If we define only two states, alive or dead, we\r
+find the usual life expectancy where it is assumed that at each age,\r
+people are at the same risk of dying. If we are differentiating the\r
+alive state into healthy and disabled, and as mortality from the\r
+disabled state is higher than mortality from the healthy state, we are\r
+introducing heterogeneity in the risk of dying. The total mortality at\r
+each age is the weighted mean of the mortality from each state by the\r
+prevalence of each state. Therefore if the proportion of people at each age and \r
+in each state is different from the period equilibrium, there is no reason to \r
+find the same total mortality at a particular age. Life expectancy, even if it \r
+is a very useful tool, has a very strong hypothesis of homogeneity of the \r
+population. Our main purpose is not to measure differential mortality but to \r
+measure the expected time in a healthy or disabled state in order to maximise \r
+the former and minimize the latter. But the differential in mortality \r
+complicates the measurement.</P>\r
+<P>Incidences of disability or recovery are not affected by the number of states \r
+if these states are independent. But incidence estimates are dependent on the \r
+specification of the model. The more covariates we add in the logit\r
+model the better \r
+is the model, but some covariates are not well measured, some are confounding \r
+factors like in any statistical model. The procedure to "fit the best model' is \r
+similar to logistic regression which itself is similar to regression analysis. \r
+We haven't yet been sofar because we also have a severe limitation which is the \r
+speed of the convergence. On a Pentium III, 500 MHz, even the simplest model, \r
+estimated by month on 8,000 people may take 4 hours to converge. Also, the IMaCh \r
+program is not a statistical package, and does not allow sophisticated design \r
+variables. If you need sophisticated design variable you have to them your self \r
+and and add them as ordinary variables. IMaCh allows up to 8 variables. The \r
+current version of this program allows only to add simple variables like age+sex \r
+or age+sex+ age*sex but will never be general enough. But what is to remember, \r
+is that incidences or probability of change from one state to another is \r
+affected by the variables specified into the model.</P>\r
+<P>Also, the age range of the people interviewed is linked the age range of the \r
+life expectancy which can be estimated by extrapolation. If your sample ranges \r
+from age 70 to 95, you can clearly estimate a life expectancy at age 70 and \r
+trust your confidence interval because it is mostly based on your sample size, \r
+but if you want to estimate the life expectancy at age 50, you should rely in \r
+the design of your model. Fitting a logistic model on a age range of 70 to 95 \r
+and estimating probabilties of transition out of this age range, say at age 50, \r
+is very dangerous. At least you should remember that the confidence interval \r
+given by the standard deviation of the health expectancies, are under the strong \r
+assumption that your model is the 'true model', which is probably not the case \r
+outside the age range of your sample.</P>\r
+<H5><FONT color=#ec5e5e size=3><B>- Copy of the parameter file</B></FONT><B>: \r
+</B><A \r
+href="http://euroreves.ined.fr/imach/doc/orbiaspar.txt"><B>orbiaspar.txt</B></A></H5>\r
+<P>This copy of the parameter file can be useful to re-run the program while \r
+saving the old output files. </P>\r
+<H5><FONT color=#ec5e5e size=3><B>- Prevalence forecasting</B></FONT><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/frbiaspar.txt"><B>biaspar/frbiaspar.txt</B></A></H5>\r
+<P>First, we have estimated the observed prevalence between 1/1/1984 and \r
+1/6/1988 (June, European syntax of dates). The mean date of all interviews \r
+(weighted average of the interviews performed between 1/1/1984 and 1/6/1988) is \r
+estimated to be 13/9/1985, as written on the top on the file. Then we forecast \r
+the probability to be in each state. </P>\r
+<P>For example on 1/1/1989 : </P><PRE class=MsoNormal># StartingAge FinalAge P.1 P.2 P.3\r
# Forecasting at date 1/1/1989\r
- 73 0.807 0.078 0.115</pre>\r
-\r
-<p>\r
-\r
-Since the minimum age is 70 on the 13/9/1985, the youngest forecasted\r
-age is 73. This means that at age a person aged 70 at 13/9/1989 has a\r
-probability to enter state1 of 0.807 at age 73 on 1/1/1989.\r
-Similarly, the probability to be in state 2 is 0.078 and the\r
-probability to die is 0.115. Then, on the 1/1/1989, the prevalence of\r
-disability at age 73 is estimated to be 0.088.</p>\r
-\r
-<h5><font color="#EC5E5E" size="3"><b>- Population forecasting</b></font><b>:\r
-</b><a href="biaspar/poprbiaspar.txt"><b>biaspar/poprbiaspar.txt</b></a></h5>\r
-\r
-<pre># Age P.1 P.2 P.3 [Population]\r
+ 73 0.807 0.078 0.115</PRE>\r
+<P>Since the minimum age is 70 on the 13/9/1985, the youngest forecasted age is \r
+73. This means that at age a person aged 70 at 13/9/1989 has a probability to \r
+enter state1 of 0.807 at age 73 on 1/1/1989. Similarly, the probability to be in \r
+state 2 is 0.078 and the probability to die is 0.115. Then, on the 1/1/1989, the \r
+prevalence of disability at age 73 is estimated to be 0.088.</P>\r
+<H5><FONT color=#ec5e5e size=3><B>- Population forecasting</B></FONT><B>: </B><A \r
+href="http://euroreves.ined.fr/imach/doc/biaspar/poprbiaspar.txt"><B>biaspar/poprbiaspar.txt</B></A></H5><PRE># Age P.1 P.2 P.3 [Population]\r
# Forecasting at date 1/1/1989 \r
75 572685.22 83798.08 \r
74 621296.51 79767.99 \r
-73 645857.70 69320.60 </pre>\r
-\r
-<pre># Forecasting at date 1/1/19909 \r
+73 645857.70 69320.60 </PRE><PRE># Forecasting at date 1/1/19909 \r
76 442986.68 92721.14 120775.48\r
75 487781.02 91367.97 121915.51\r
-74 512892.07 85003.47 117282.76 </pre>\r
-\r
-<p>From the population file, we estimate the number of people in\r
-each state. At age 73, 645857 persons are in state 1 and 69320\r
-are in state 2. One year latter, 512892 are still in state 1,\r
-85003 are in state 2 and 117282 died before 1/1/1990.</p>\r
-\r
-<hr>\r
-\r
-<h2><a name="example"></a><font color="#00006A">Trying an example</font></h2>\r
-\r
-<p>Since you know how to run the program, it is time to test it\r
-on your own computer. Try for example on a parameter file named <a\r
-href="imachpar.imach">imachpar.imach</a> which is a copy\r
-of <font size="2" face="Courier New">mypar.imach</font> included\r
-in the subdirectory of imach, <font size="2" face="Courier New">mytry</font>.\r
-Edit it and change the name of the data file to <font size="2"\r
-face="Courier New">mydata.txt</font> if you don't want to\r
-copy it on the same directory. The file <font face="Courier New">mydata.txt</font>\r
-is a smaller file of 3,000 people but still with 4 waves. </p>\r
-\r
-<p>Right click on the .imach file and a window will popup with the\r
-string '<strong>Enter the parameter file name:'</strong></p>\r
-\r
-<table border="1">\r
- <tr>\r
- <td width="100%"><strong>IMACH, Version 0.97b</strong><p><strong>Enter\r
- the parameter file name: imachpar.imach</strong></p>\r
- </td>\r
- </tr>\r
-</table>\r
-\r
-<p>Most of the data files or image files generated, will use the\r
-'imachpar' string into their name. The running time is about 2-3\r
-minutes on a Pentium III. If the execution worked correctly, the\r
-outputs files are created in the current directory, and should be\r
-the same as the mypar files initially included in the directory <font\r
-size="2" face="Courier New">mytry</font>.</p>\r
-\r
-<ul>\r
- <li><pre><u>Output on the screen</u> The output screen looks like <a\r
-href="biaspar.log">biaspar.log</a>\r
+74 512892.07 85003.47 117282.76 </PRE>\r
+<P>From the population file, we estimate the number of people in each state. At \r
+age 73, 645857 persons are in state 1 and 69320 are in state 2. One year latter, \r
+512892 are still in state 1, 85003 are in state 2 and 117282 died before \r
+1/1/1990.</P>\r
+<HR>\r
+\r
+<H2><A name=example></A><FONT color=#00006a>Trying an example</FONT></H2>\r
+<P>Since you know how to run the program, it is time to test it on your own \r
+computer. Try for example on a parameter file named <A \r
+href="http://euroreves.ined.fr/imach/doc/imachpar.imach">imachpar.imach</A> \r
+which is a copy of <FONT face="Courier New" size=2>mypar.imach</FONT> included \r
+in the subdirectory of imach, <FONT face="Courier New" size=2>mytry</FONT>. Edit \r
+it and change the name of the data file to <FONT face="Courier New" \r
+size=2>mydata.txt</FONT> if you don't want to copy it on the same directory. The \r
+file <FONT face="Courier New">mydata.txt</FONT> is a smaller file of 3,000 \r
+people but still with 4 waves. </P>\r
+<P>Right click on the .imach file and a window will popup with the string \r
+'<STRONG>Enter the parameter file name:'</STRONG></P>\r
+<TABLE border=1>\r
+ <TBODY>\r
+ <TR>\r
+ <TD width="100%"><STRONG>IMACH, Version 0.97b</STRONG>\r
+ <P><STRONG>Enter the parameter file name: \r
+ imachpar.imach</STRONG></P></TD></TR></TBODY></TABLE>\r
+<P>Most of the data files or image files generated, will use the 'imachpar' \r
+string into their name. The running time is about 2-3 minutes on a Pentium III. \r
+If the execution worked correctly, the outputs files are created in the current \r
+directory, and should be the same as the mypar files initially included in the \r
+directory <FONT face="Courier New" size=2>mytry</FONT>.</P>\r
+<UL>\r
+ <LI><PRE><U>Output on the screen</U> The output screen looks like <A href="http://euroreves.ined.fr/imach/doc/biaspar.log">biaspar.log</A>\r
#\r
title=MLE datafile=mydaiata.txt lastobs=3000 firstpass=1 lastpass=3\r
-ftol=1.000000e-008 stepm=24 ncovcol=2 nlstate=2 ndeath=1 maxwav=4 mle=1 weight=0</pre>\r
- </li>\r
- <li><pre>Total number of individuals= 2965, Agemin = 70.00, Agemax= 100.92\r
+ftol=1.000000e-008 stepm=24 ncovcol=2 nlstate=2 ndeath=1 maxwav=4 mle=1 weight=0</PRE>\r
+ <LI><PRE>Total number of individuals= 2965, Agemin = 70.00, Agemax= 100.92\r
\r
Warning, no any valid information for:126 line=126\r
Warning, no any valid information for:2307 line=2307\r
Delay (in months) between two waves Min=21 Max=51 Mean=24.495826\r
-<font face="Times New Roman">These lines give some warnings on the data file and also some raw statistics on frequencies of transitions.</font>\r
+<FONT face="Times New Roman">These lines give some warnings on the data file and also some raw statistics on frequencies of transitions.</FONT>\r
Age 70 1.=230 loss[1]=3.5% 2.=16 loss[2]=12.5% 1.=222 prev[1]=94.1% 2.=14\r
prev[2]=5.9% 1-1=8 11=200 12=7 13=15 2-1=2 21=6 22=7 23=1\r
-Age 102 1.=0 loss[1]=NaNQ% 2.=0 loss[2]=NaNQ% 1.=0 prev[1]=NaNQ% 2.=0 </pre>\r
- </li>\r
-</ul>\r
-It includes some warnings or errors which are very important for\r
-you. Be careful with such warnings because your results may be biased\r
-if, for example, you have people who accepted to be interviewed at\r
-first pass but never after. Or if you don't have the exact month of\r
-death. In such cases IMaCh doesn't take any initiative, it does only\r
-warn you. It is up to you to decide what to do with these\r
-people. Excluding them is usually a wrong decision. It is better to\r
-decide that the month of death is at the mid-interval between the last\r
-two waves for example.<p>\r
-\r
-If you survey suffers from severe attrition, you have to analyse the\r
-characteristics of the lost people and overweight people with same\r
-characteristics for example.\r
-<p>\r
-By default, IMaCH warns and excludes these problematic people, but you\r
-have to be careful with such results.\r
-\r
-<p> </p>\r
-\r
-<ul>\r
- <li>Maximisation with the Powell algorithm. 8 directions are\r
- given corresponding to the 8 parameters. this can be\r
- rather long to get convergence.<br>\r
- <font size="1" face="Courier New"><br>\r
- Powell iter=1 -2*LL=11531.405658264877 1 0.000000000000 2\r
- 0.000000000000 3<br>\r
- 0.000000000000 4 0.000000000000 5 0.000000000000 6\r
- 0.000000000000 7 <br>\r
- 0.000000000000 8 0.000000000000<br>\r
- 1..........2.................3..........4.................5.........<br>\r
- 6................7........8...............<br>\r
- Powell iter=23 -2*LL=6744.954108371555 1 -12.967632334283\r
- <br>\r
- 2 0.135136681033 3 -7.402109728262 4 0.067844593326 <br>\r
- 5 -0.673601538129 6 -0.006615504377 7 -5.051341616718 <br>\r
- 8 0.051272038506<br>\r
- 1..............2...........3..............4...........<br>\r
- 5..........6................7...........8.........<br>\r
- #Number of iterations = 23, -2 Log likelihood =\r
- 6744.954042573691<br>\r
- # Parameters<br>\r
- 12 -12.966061 0.135117 <br>\r
- 13 -7.401109 0.067831 <br>\r
- 21 -0.672648 -0.006627 <br>\r
- 23 -5.051297 0.051271 </font><br>\r
- </li>\r
- <li><pre><font size="2">Calculation of the hessian matrix. Wait...\r
+Age 102 1.=0 loss[1]=NaNQ% 2.=0 loss[2]=NaNQ% 1.=0 prev[1]=NaNQ% 2.=0 </PRE></LI></UL>It \r
+includes some warnings or errors which are very important for you. Be careful \r
+with such warnings because your results may be biased if, for example, you have \r
+people who accepted to be interviewed at first pass but never after. Or if you \r
+don't have the exact month of death. In such cases IMaCh doesn't take any \r
+initiative, it does only warn you. It is up to you to decide what to do with \r
+these people. Excluding them is usually a wrong decision. It is better to decide \r
+that the month of death is at the mid-interval between the last two waves for \r
+example.\r
+<P>If you survey suffers from severe attrition, you have to analyse the \r
+characteristics of the lost people and overweight people with same \r
+characteristics for example. \r
+<P>By default, IMaCH warns and excludes these problematic people, but you have \r
+to be careful with such results. \r
+<P> </P>\r
+<UL>\r
+ <LI>Maximisation with the Powell algorithm. 8 directions are given \r
+ corresponding to the 8 parameters. this can be rather long to get \r
+ convergence.<BR><FONT face="Courier New" size=1><BR>Powell iter=1 \r
+ -2*LL=11531.405658264877 1 0.000000000000 2 0.000000000000 3<BR>0.000000000000 \r
+ 4 0.000000000000 5 0.000000000000 6 0.000000000000 7 <BR>0.000000000000 8 \r
+ 0.000000000000<BR>1..........2.................3..........4.................5.........<BR>6................7........8...............<BR>Powell \r
+ iter=23 -2*LL=6744.954108371555 1 -12.967632334283 <BR>2 0.135136681033 3 \r
+ -7.402109728262 4 0.067844593326 <BR>5 -0.673601538129 6 -0.006615504377 7 \r
+ -5.051341616718 <BR>8 \r
+ 0.051272038506<BR>1..............2...........3..............4...........<BR>5..........6................7...........8.........<BR>#Number \r
+ of iterations = 23, -2 Log likelihood = 6744.954042573691<BR># \r
+ Parameters<BR>12 -12.966061 0.135117 <BR>13 -7.401109 0.067831 <BR>21 \r
+ -0.672648 -0.006627 <BR>23 -5.051297 0.051271 </FONT><BR>\r
+ <LI><PRE><FONT size=2>Calculation of the hessian matrix. Wait...\r
12345678.12.13.14.15.16.17.18.23.24.25.26.27.28.34.35.36.37.38.45.46.47.48.56.57.58.67.68.78\r
\r
Inverting the hessian to get the covariance matrix. Wait...\r
Computing Total LEs with variances: file 'trmypar.txt' \r
Computing Variance-covariance of Prevalence limit: file 'vplrmypar.txt' \r
End of Imach\r
-</font></pre>\r
- </li>\r
-</ul>\r
-\r
-<p><font size="3">Once the running is finished, the program\r
-requires a character:</font></p>\r
-\r
-<table border="1">\r
- <tr>\r
- <td width="100%"><strong>Type e to edit output files, g\r
- to graph again, c to start again, and q for exiting:</strong></td>\r
- </tr>\r
-</table>\r
-\r
-In order to have an idea of the time needed to reach convergence,\r
-IMaCh gives an estimation if the convergence needs 10, 20 or 30\r
-iterations. It might be useful.\r
-\r
-<p><font size="3">First you should enter <strong>e </strong>to\r
-edit the master file mypar.htm. </font></p>\r
-\r
-<ul>\r
- <li><u>Outputs files</u> <br>\r
- <br>\r
- - Copy of the parameter file: <a href="ormypar.txt">ormypar.txt</a><br>\r
- - Gnuplot file name: <a href="mypar.gp.txt">mypar.gp.txt</a><br>\r
- - Cross-sectional prevalence in each state: <a\r
- href="prmypar.txt">prmypar.txt</a> <br>\r
- - Period prevalence in each state: <a\r
- href="plrmypar.txt">plrmypar.txt</a> <br>\r
- - Transition probabilities: <a href="pijrmypar.txt">pijrmypar.txt</a><br>\r
- - Life expectancies by age and initial health status\r
- (estepm=24 months): <a href="ermypar.txt">ermypar.txt</a>\r
- <br>\r
- - Parameter file with estimated parameters and the\r
- covariance matrix: <a href="rmypar.txt">rmypar.txt</a> <br>\r
- - Variance of one-step probabilities: <a\r
- href="probrmypar.txt">probrmypar.txt</a> <br>\r
- - Variances of life expectancies by age and initial\r
- health status (estepm=24 months): <a href="vrmypar.txt">vrmypar.txt</a><br>\r
- - Health expectancies with their variances: <a\r
- href="trmypar.txt">trmypar.txt</a> <br>\r
- - Standard deviation of period prevalences: <a\r
- href="vplrmypar.txt">vplrmypar.txt</a> <br>\r
- No population forecast: popforecast = 0 (instead of 1) or\r
- stepm = 24 (instead of 1) or model=. (instead of .)<br>\r
- <br>\r
- </li>\r
- <li><u>Graphs</u> <br>\r
- <br>\r
- -<a href="../mytry/pemypar1.gif">One-step transition\r
- probabilities</a><br>\r
- -<a href="../mytry/pmypar11.gif">Convergence to the\r
- period prevalence</a><br>\r
- -<a href="..\mytry\vmypar11.gif">Cross-sectional and period\r
- prevalence in state (1) with the confident interval</a> <br>\r
- -<a href="..\mytry\vmypar21.gif">Cross-sectional and period\r
- prevalence in state (2) with the confident interval</a> <br>\r
- -<a href="..\mytry\expmypar11.gif">Health life\r
- expectancies by age and initial health state (1)</a> <br>\r
- -<a href="..\mytry\expmypar21.gif">Health life\r
- expectancies by age and initial health state (2)</a> <br>\r
- -<a href="..\mytry\emypar1.gif">Total life expectancy by\r
- age and health expectancies in states (1) and (2).</a> </li>\r
-</ul>\r
-\r
-<p>This software have been partly granted by <a\r
-href="http://euroreves.ined.fr">Euro-REVES</a>, a concerted action\r
-from the European Union. Since 2003 it is also partly granted by the\r
-French Institute on Longevity. It will be copyrighted identically to a\r
-GNU software product, i.e. program and software can be distributed\r
-freely for non commercial use. Sources are not widely distributed\r
-today because some part of the codes are copyrighted by Numerical\r
-Recipes in C. You can get our GPL codes by asking us with a simple\r
-justification (name, email, institute) <a\r
-href="mailto:brouard@ined.fr">mailto:brouard@ined.fr</a> and <a\r
-href="mailto:lievre@ined.fr">mailto:lievre@ined.fr</a> .</p>\r
-\r
-<p>Latest version (0.97b of June 2004) can be accessed at <a\r
-href="http://euroreves.ined.fr/imach">http://euroreves.ined.fr/imach</a><br>\r
-</p>\r
-</body>\r
-</html>\r
+</FONT></PRE></LI></UL>\r
+<P><FONT size=3>Once the running is finished, the program requires a \r
+character:</FONT></P>\r
+<TABLE border=1>\r
+ <TBODY>\r
+ <TR>\r
+ <TD width="100%"><STRONG>Type e to edit output files, g to graph again, c \r
+ to start again, and q for exiting:</STRONG></TD></TR></TBODY></TABLE>In order to \r
+have an idea of the time needed to reach convergence, IMaCh gives an estimation \r
+if the convergence needs 10, 20 or 30 iterations. It might be useful. \r
+<P><FONT size=3>First you should enter <STRONG>e </STRONG>to edit the master \r
+file mypar.htm. </FONT></P>\r
+<UL>\r
+ <LI><U>Outputs files</U> <BR><BR>- Copy of the parameter file: <A \r
+ href="http://euroreves.ined.fr/imach/doc/ormypar.txt">ormypar.txt</A><BR>- \r
+ Gnuplot file name: <A \r
+ href="http://euroreves.ined.fr/imach/doc/mypar.gp.txt">mypar.gp.txt</A><BR>- \r
+ Cross-sectional prevalence in each state: <A \r
+ href="http://euroreves.ined.fr/imach/doc/prmypar.txt">prmypar.txt</A> <BR>- \r
+ Period prevalence in each state: <A \r
+ href="http://euroreves.ined.fr/imach/doc/plrmypar.txt">plrmypar.txt</A> <BR>- \r
+ Transition probabilities: <A \r
+ href="http://euroreves.ined.fr/imach/doc/pijrmypar.txt">pijrmypar.txt</A><BR>- \r
+ Life expectancies by age and initial health status (estepm=24 months): <A \r
+ href="http://euroreves.ined.fr/imach/doc/ermypar.txt">ermypar.txt</A> <BR>- \r
+ Parameter file with estimated parameters and the covariance matrix: <A \r
+ href="http://euroreves.ined.fr/imach/doc/rmypar.txt">rmypar.txt</A> <BR>- \r
+ Variance of one-step probabilities: <A \r
+ href="http://euroreves.ined.fr/imach/doc/probrmypar.txt">probrmypar.txt</A> \r
+ <BR>- Variances of life expectancies by age and initial health status \r
+ (estepm=24 months): <A \r
+ href="http://euroreves.ined.fr/imach/doc/vrmypar.txt">vrmypar.txt</A><BR>- \r
+ Health expectancies with their variances: <A \r
+ href="http://euroreves.ined.fr/imach/doc/trmypar.txt">trmypar.txt</A> <BR>- \r
+ Standard deviation of period prevalences: <A \r
+ href="http://euroreves.ined.fr/imach/doc/vplrmypar.txt">vplrmypar.txt</A> \r
+ <BR>No population forecast: popforecast = 0 (instead of 1) or stepm = 24 \r
+ (instead of 1) or model=. (instead of .)<BR><BR>\r
+ <LI><U>Graphs</U> <BR><BR>-<A \r
+ href="http://euroreves.ined.fr/imach/mytry/pemypar1.gif">One-step transition \r
+ probabilities</A><BR>-<A \r
+ href="http://euroreves.ined.fr/imach/mytry/pmypar11.gif">Convergence to the \r
+ period prevalence</A><BR>-<A \r
+ href="http://euroreves.ined.fr/imach/mytry/vmypar11.gif">Cross-sectional and \r
+ period prevalence in state (1) with the confident interval</A> <BR>-<A \r
+ href="http://euroreves.ined.fr/imach/mytry/vmypar21.gif">Cross-sectional and \r
+ period prevalence in state (2) with the confident interval</A> <BR>-<A \r
+ href="http://euroreves.ined.fr/imach/mytry/expmypar11.gif">Health life \r
+ expectancies by age and initial health state (1)</A> <BR>-<A \r
+ href="http://euroreves.ined.fr/imach/mytry/expmypar21.gif">Health life \r
+ expectancies by age and initial health state (2)</A> <BR>-<A \r
+ href="http://euroreves.ined.fr/imach/mytry/emypar1.gif">Total life expectancy \r
+ by age and health expectancies in states (1) and (2).</A> </LI></UL>\r
+<P>This software have been partly granted by <A \r
+href="http://euroreves.ined.fr/">Euro-REVES</A>, a concerted action from the \r
+European Union. It will be copyrighted identically to a GNU software product, \r
+i.e. program and software can be distributed freely for non commercial use. \r
+Sources are not widely distributed today. You can get them by asking us with a \r
+simple justification (name, email, institute) <A \r
+href="mailto:brouard@ined.fr">mailto:brouard@ined.fr</A> and <A \r
+href="mailto:lievre@ined.fr">mailto:lievre@ined.fr</A> .</P>\r
+<P>Latest version (0.97b of June 2004) can be accessed at <A \r
+href="http://euroreves.ined.fr/imach">http://euroreves.ined.fr/imach</A><BR></P></BODY></HTML>\r
git://henry.ined.fr / .git / commitdiff