Annotation of imach096d/doc/imach.htm, revision 1.9
1.9 ! brouard 1: <!-- $Id: imach.htm,v 1.8 2002/03/11 14:18:06 lievre Exp $ --!>
1.2 lievre 2: <html>
3:
4: <head>
5: <meta http-equiv="Content-Type"
6: content="text/html; charset=iso-8859-1">
7: <meta name="GENERATOR" content="Microsoft FrontPage Express 2.0">
8: <title>Computing Health Expectancies using IMaCh</title>
9: <!-- Changed by: Agnes Lievre, 12-Oct-2000 -->
1.8 lievre 10: <html>
11:
12: <head>
13: <meta http-equiv="Content-Type"
14: content="text/html; charset=iso-8859-1">
15: <meta name="GENERATOR" content="Microsoft FrontPage Express 2.0">
16: <title></title>
1.2 lievre 17: </head>
18:
1.6 lievre 19: <body bgcolor="#FFFFFF">
1.2 lievre 20:
1.6 lievre 21: <hr size="3" color="#EC5E5E">
1.2 lievre 22:
1.6 lievre 23: <h1 align="center"><font color="#00006A">Computing Health
24: Expectancies using IMaCh</font></h1>
1.2 lievre 25:
1.6 lievre 26: <h1 align="center"><font color="#00006A" size="5">(a Maximum
27: Likelihood Computer Program using Interpolation of Markov Chains)</font></h1>
28:
29: <p align="center"> </p>
30:
31: <p align="center"><a href="http://www.ined.fr/"><img
32: src="logo-ined.gif" border="0" width="151" height="76"></a><img
33: src="euroreves2.gif" width="151" height="75"></p>
34:
35: <h3 align="center"><a href="http://www.ined.fr/"><font
36: color="#00006A">INED</font></a><font color="#00006A"> and </font><a
37: href="http://euroreves.ined.fr"><font color="#00006A">EUROREVES</font></a></h3>
38:
39: <p align="center"><font color="#00006A" size="4"><strong>Version
1.7 brouard 40: 0.71a, March 2002</strong></font></p>
1.6 lievre 41:
42: <hr size="3" color="#EC5E5E">
43:
44: <p align="center"><font color="#00006A"><strong>Authors of the
45: program: </strong></font><a href="http://sauvy.ined.fr/brouard"><font
46: color="#00006A"><strong>Nicolas Brouard</strong></font></a><font
47: color="#00006A"><strong>, senior researcher at the </strong></font><a
48: href="http://www.ined.fr"><font color="#00006A"><strong>Institut
49: National d'Etudes Démographiques</strong></font></a><font
50: color="#00006A"><strong> (INED, Paris) in the "Mortality,
51: Health and Epidemiology" Research Unit </strong></font></p>
52:
53: <p align="center"><font color="#00006A"><strong>and Agnès
54: Lièvre<br clear="left">
55: </strong></font></p>
56:
57: <h4><font color="#00006A">Contribution to the mathematics: C. R.
58: Heathcote </font><font color="#00006A" size="2">(Australian
59: National University, Canberra).</font></h4>
60:
61: <h4><font color="#00006A">Contact: Agnès Lièvre (</font><a
62: href="mailto:lievre@ined.fr"><font color="#00006A"><i>lievre@ined.fr</i></font></a><font
63: color="#00006A">) </font></h4>
1.2 lievre 64:
65: <hr>
1.6 lievre 66:
67: <ul>
68: <li><a href="#intro">Introduction</a> </li>
69: <li><a href="#data">On what kind of data can it be used?</a></li>
70: <li><a href="#datafile">The data file</a> </li>
71: <li><a href="#biaspar">The parameter file</a> </li>
72: <li><a href="#running">Running Imach</a> </li>
73: <li><a href="#output">Output files and graphs</a> </li>
74: <li><a href="#example">Exemple</a> </li>
1.2 lievre 75: </ul>
1.6 lievre 76:
1.2 lievre 77: <hr>
78:
1.6 lievre 79: <h2><a name="intro"><font color="#00006A">Introduction</font></a></h2>
1.2 lievre 80:
1.6 lievre 81: <p>This program computes <b>Healthy Life Expectancies</b> from <b>cross-longitudinal
82: data</b> using the methodology pioneered by Laditka and Wolf (1).
83: Within the family of Health Expectancies (HE), Disability-free
84: life expectancy (DFLE) is probably the most important index to
1.2 lievre 85: monitor. In low mortality countries, there is a fear that when
86: mortality declines, the increase in DFLE is not proportionate to
87: the increase in total Life expectancy. This case is called the <em>Expansion
88: of morbidity</em>. Most of the data collected today, in
1.7 brouard 89: particular by the international <a href="http://www.reves.org">REVES</a>
1.2 lievre 90: network on Health expectancy, and most HE indices based on these
91: data, are <em>cross-sectional</em>. It means that the information
92: collected comes from a single cross-sectional survey: people from
93: various ages (but mostly old people) are surveyed on their health
94: status at a single date. Proportion of people disabled at each
95: age, can then be measured at that date. This age-specific
96: prevalence curve is then used to distinguish, within the
97: stationary population (which, by definition, is the life table
98: estimated from the vital statistics on mortality at the same
99: date), the disable population from the disability-free
100: population. Life expectancy (LE) (or total population divided by
101: the yearly number of births or deaths of this stationary
102: population) is then decomposed into DFLE and DLE. This method of
103: computing HE is usually called the Sullivan method (from the name
1.6 lievre 104: of the author who first described it).</p>
1.2 lievre 105:
1.6 lievre 106: <p>Age-specific proportions of people disable are very difficult
107: to forecast because each proportion corresponds to historical
108: conditions of the cohort and it is the result of the historical
109: flows from entering disability and recovering in the past until
110: today. The age-specific intensities (or incidence rates) of
111: entering disability or recovering a good health, are reflecting
112: actual conditions and therefore can be used at each age to
113: forecast the future of this cohort. For example if a country is
114: improving its technology of prosthesis, the incidence of
115: recovering the ability to walk will be higher at each (old) age,
116: but the prevalence of disability will only slightly reflect an
117: improve because the prevalence is mostly affected by the history
118: of the cohort and not by recent period effects. To measure the
119: period improvement we have to simulate the future of a cohort of
120: new-borns entering or leaving at each age the disability state or
121: dying according to the incidence rates measured today on
122: different cohorts. The proportion of people disabled at each age
123: in this simulated cohort will be much lower (using the exemple of
124: an improvement) that the proportions observed at each age in a
125: cross-sectional survey. This new prevalence curve introduced in a
126: life table will give a much more actual and realistic HE level
127: than the Sullivan method which mostly measured the History of
128: health conditions in this country.</p>
129:
130: <p>Therefore, the main question is how to measure incidence rates
131: from cross-longitudinal surveys? This is the goal of the IMaCH
132: program. From your data and using IMaCH you can estimate period
133: HE and not only Sullivan's HE. Also the standard errors of the HE
134: are computed.</p>
135:
136: <p>A cross-longitudinal survey consists in a first survey
137: ("cross") where individuals from different ages are
138: interviewed on their health status or degree of disability. At
139: least a second wave of interviews ("longitudinal")
140: should measure each new individual health status. Health
141: expectancies are computed from the transitions observed between
142: waves and are computed for each degree of severity of disability
143: (number of life states). More degrees you consider, more time is
144: necessary to reach the Maximum Likelihood of the parameters
145: involved in the model. Considering only two states of disability
146: (disable and healthy) is generally enough but the computer
147: program works also with more health statuses.<br>
1.2 lievre 148: <br>
149: The simplest model is the multinomial logistic model where <i>pij</i>
150: is the probability to be observed in state <i>j</i> at the second
151: wave conditional to be observed in state <em>i</em> at the first
152: wave. Therefore a simple model is: log<em>(pij/pii)= aij +
153: bij*age+ cij*sex,</em> where '<i>age</i>' is age and '<i>sex</i>'
154: is a covariate. The advantage that this computer program claims,
155: comes from that if the delay between waves is not identical for
156: each individual, or if some individual missed an interview, the
157: information is not rounded or lost, but taken into account using
158: an interpolation or extrapolation. <i>hPijx</i> is the
159: probability to be observed in state <i>i</i> at age <i>x+h</i>
160: conditional to the observed state <i>i</i> at age <i>x</i>. The
161: delay '<i>h</i>' can be split into an exact number (<i>nh*stepm</i>)
162: of unobserved intermediate states. This elementary transition (by
163: month or quarter trimester, semester or year) is modeled as a
164: multinomial logistic. The <i>hPx</i> matrix is simply the matrix
165: product of <i>nh*stepm</i> elementary matrices and the
166: contribution of each individual to the likelihood is simply <i>hPijx</i>.
1.6 lievre 167: <br>
168: </p>
1.2 lievre 169:
1.6 lievre 170: <p>The program presented in this manual is a quite general
171: program named <strong>IMaCh</strong> (for <strong>I</strong>nterpolated
172: <strong>MA</strong>rkov <strong>CH</strong>ain), designed to
173: analyse transition data from longitudinal surveys. The first step
174: is the parameters estimation of a transition probabilities model
175: between an initial status and a final status. From there, the
176: computer program produces some indicators such as observed and
177: stationary prevalence, life expectancies and their variances and
178: graphs. Our transition model consists in absorbing and
179: non-absorbing states with the possibility of return across the
180: non-absorbing states. The main advantage of this package,
1.2 lievre 181: compared to other programs for the analysis of transition data
1.6 lievre 182: (For example: Proc Catmod of SAS<sup>®</sup>) is that the whole
1.2 lievre 183: individual information is used even if an interview is missing, a
184: status or a date is unknown or when the delay between waves is
185: not identical for each individual. The program can be executed
186: according to parameters: selection of a sub-sample, number of
187: absorbing and non-absorbing states, number of waves taken in
188: account (the user inputs the first and the last interview), a
189: tolerance level for the maximization function, the periodicity of
1.5 lievre 190: the transitions (we can compute annual, quarterly or monthly
1.2 lievre 191: transitions), covariates in the model. It works on Windows or on
1.6 lievre 192: Unix.<br>
193: </p>
1.2 lievre 194:
195: <hr>
196:
1.6 lievre 197: <p>(1) Laditka, Sarah B. and Wolf, Douglas A. (1998), "New
1.2 lievre 198: Methods for Analyzing Active Life Expectancy". <i>Journal of
1.6 lievre 199: Aging and Health</i>. Vol 10, No. 2. </p>
1.2 lievre 200:
201: <hr>
202:
1.6 lievre 203: <h2><a name="data"><font color="#00006A">On what kind of data can
204: it be used?</font></a></h2>
1.2 lievre 205:
1.6 lievre 206: <p>The minimum data required for a transition model is the
207: recording of a set of individuals interviewed at a first date and
208: interviewed again at least one another time. From the
209: observations of an individual, we obtain a follow-up over time of
210: the occurrence of a specific event. In this documentation, the
211: event is related to health status at older ages, but the program
212: can be applied on a lot of longitudinal studies in different
213: contexts. To build the data file explained into the next section,
214: you must have the month and year of each interview and the
215: corresponding health status. But in order to get age, date of
216: birth (month and year) is required (missing values is allowed for
217: month). Date of death (month and year) is an important
218: information also required if the individual is dead. Shorter
219: steps (i.e. a month) will more closely take into account the
220: survival time after the last interview.</p>
1.2 lievre 221:
222: <hr>
223:
1.6 lievre 224: <h2><a name="datafile"><font color="#00006A">The data file</font></a></h2>
1.2 lievre 225:
1.6 lievre 226: <p>In this example, 8,000 people have been interviewed in a
227: cross-longitudinal survey of 4 waves (1984, 1986, 1988, 1990).
228: Some people missed 1, 2 or 3 interviews. Health statuses are
229: healthy (1) and disable (2). The survey is not a real one. It is
230: a simulation of the American Longitudinal Survey on Aging. The
231: disability state is defined if the individual missed one of four
232: ADL (Activity of daily living, like bathing, eating, walking).
233: Therefore, even is the individuals interviewed in the sample are
234: virtual, the information brought with this sample is close to the
235: situation of the United States. Sex is not recorded is this
236: sample.</p>
237:
238: <p>Each line of the data set (named <a href="data1.txt">data1.txt</a>
239: in this first example) is an individual record which fields are: </p>
240:
241: <ul>
242: <li><b>Index number</b>: positive number (field 1) </li>
243: <li><b>First covariate</b> positive number (field 2) </li>
244: <li><b>Second covariate</b> positive number (field 3) </li>
245: <li><a name="Weight"><b>Weight</b></a>: positive number
246: (field 4) . In most surveys individuals are weighted
247: according to the stratification of the sample.</li>
248: <li><b>Date of birth</b>: coded as mm/yyyy. Missing dates are
249: coded as 99/9999 (field 5) </li>
250: <li><b>Date of death</b>: coded as mm/yyyy. Missing dates are
251: coded as 99/9999 (field 6) </li>
252: <li><b>Date of first interview</b>: coded as mm/yyyy. Missing
253: dates are coded as 99/9999 (field 7) </li>
254: <li><b>Status at first interview</b>: positive number.
255: Missing values ar coded -1. (field 8) </li>
256: <li><b>Date of second interview</b>: coded as mm/yyyy.
257: Missing dates are coded as 99/9999 (field 9) </li>
258: <li><strong>Status at second interview</strong> positive
259: number. Missing values ar coded -1. (field 10) </li>
260: <li><b>Date of third interview</b>: coded as mm/yyyy. Missing
261: dates are coded as 99/9999 (field 11) </li>
262: <li><strong>Status at third interview</strong> positive
263: number. Missing values ar coded -1. (field 12) </li>
264: <li><b>Date of fourth interview</b>: coded as mm/yyyy.
265: Missing dates are coded as 99/9999 (field 13) </li>
266: <li><strong>Status at fourth interview</strong> positive
267: number. Missing values are coded -1. (field 14) </li>
268: <li>etc</li>
1.2 lievre 269: </ul>
270:
1.6 lievre 271: <p> </p>
1.2 lievre 272:
1.6 lievre 273: <p>If your longitudinal survey do not include information about
274: weights or covariates, you must fill the column with a number
275: (e.g. 1) because a missing field is not allowed.</p>
1.2 lievre 276:
277: <hr>
278:
1.6 lievre 279: <h2><font color="#00006A">Your first example parameter file</font><a
280: href="http://euroreves.ined.fr/imach"></a><a name="uio"></a></h2>
1.2 lievre 281:
1.7 brouard 282: <h2><a name="biaspar"></a>#Imach version 0.71a, March 2002,
1.6 lievre 283: INED-EUROREVES </h2>
1.2 lievre 284:
1.6 lievre 285: <p>This is a comment. Comments start with a '#'.</p>
1.2 lievre 286:
1.6 lievre 287: <h4><font color="#FF0000">First uncommented line</font></h4>
288:
289: <pre>title=1st_example datafile=data1.txt lastobs=8600 firstpass=1 lastpass=4</pre>
290:
291: <ul>
292: <li><b>title=</b> 1st_example is title of the run. </li>
293: <li><b>datafile=</b>data1.txt is the name of the data set.
294: Our example is a six years follow-up survey. It consists
295: in a baseline followed by 3 reinterviews. </li>
296: <li><b>lastobs=</b> 8600 the program is able to run on a
297: subsample where the last observation number is lastobs.
298: It can be set a bigger number than the real number of
299: observations (e.g. 100000). In this example, maximisation
300: will be done on the 8600 first records. </li>
301: <li><b>firstpass=1</b> , <b>lastpass=4 </b>In case of more
302: than two interviews in the survey, the program can be run
303: on selected transitions periods. firstpass=1 means the
304: first interview included in the calculation is the
305: baseline survey. lastpass=4 means that the information
306: brought by the 4th interview is taken into account.</li>
1.2 lievre 307: </ul>
308:
1.6 lievre 309: <p> </p>
310:
311: <h4><a name="biaspar-2"><font color="#FF0000">Second uncommented
312: line</font></a></h4>
1.2 lievre 313:
1.6 lievre 314: <pre>ftol=1.e-08 stepm=1 ncov=2 nlstate=2 ndeath=1 maxwav=4 mle=1 weight=0</pre>
315:
316: <ul>
317: <li><b>ftol=1e-8</b> Convergence tolerance on the function
318: value in the maximisation of the likelihood. Choosing a
319: correct value for ftol is difficult. 1e-8 is a correct
320: value for a 32 bits computer.</li>
321: <li><b>stepm=1</b> Time unit in months for interpolation.
322: Examples:<ul>
323: <li>If stepm=1, the unit is a month </li>
324: <li>If stepm=4, the unit is a trimester</li>
325: <li>If stepm=12, the unit is a year </li>
326: <li>If stepm=24, the unit is two years</li>
327: <li>... </li>
1.2 lievre 328: </ul>
329: </li>
1.8 lievre 330: <li><b>ncov=2</b> Number of covariates in the datafile. </li>
1.6 lievre 331: <li><b>nlstate=2</b> Number of non-absorbing (alive) states.
332: Here we have two alive states: disability-free is coded 1
333: and disability is coded 2. </li>
334: <li><b>ndeath=1</b> Number of absorbing states. The absorbing
335: state death is coded 3. </li>
336: <li><b>maxwav=4</b> Number of waves in the datafile.</li>
337: <li><a name="mle"><b>mle</b></a><b>=1</b> Option for the
338: Maximisation Likelihood Estimation. <ul>
339: <li>If mle=1 the program does the maximisation and
340: the calculation of health expectancies </li>
341: <li>If mle=0 the program only does the calculation of
342: the health expectancies. </li>
1.2 lievre 343: </ul>
344: </li>
1.6 lievre 345: <li><b>weight=0</b> Possibility to add weights. <ul>
346: <li>If weight=0 no weights are included </li>
347: <li>If weight=1 the maximisation integrates the
348: weights which are in field <a href="#Weight">4</a></li>
1.2 lievre 349: </ul>
350: </li>
351: </ul>
352:
1.6 lievre 353: <h4><font color="#FF0000">Covariates</font></h4>
354:
355: <p>Intercept and age are systematically included in the model.
1.8 lievre 356: Additional covariates can be included with the command: </p>
1.2 lievre 357:
1.6 lievre 358: <pre>model=<em>list of covariates</em></pre>
359:
360: <ul>
361: <li>if<strong> model=. </strong>then no covariates are
362: included</li>
363: <li>if <strong>model=V1</strong> the model includes the first
364: covariate (field 2)</li>
365: <li>if <strong>model=V2 </strong>the model includes the
366: second covariate (field 3)</li>
367: <li>if <strong>model=V1+V2 </strong>the model includes the
368: first and the second covariate (fields 2 and 3)</li>
369: <li>if <strong>model=V1*V2 </strong>the model includes the
1.2 lievre 370: product of the first and the second covariate (fields 2
1.6 lievre 371: and 3)</li>
372: <li>if <strong>model=V1+V1*age</strong> the model includes
373: the product covariate*age</li>
1.2 lievre 374: </ul>
375:
1.8 lievre 376: <p>In this example, we have two covariates in the data file
377: (fields 2 and 3). The number of covariates is defined with
378: statement ncov=2. If now you have 3 covariates in the datafile
379: (fields 2, 3 and 4), you have to set ncov=3. Then you can run the
380: programme with a new parametrisation taking into account the
381: third covariate. For example, <strong>model=V1+V3 </strong>estimates
382: a model with the first and third covariates. More complicated
383: models can be used, but it will takes more time to converge. With
384: a simple model (no covariates), the programme estimates 8
385: parameters. Adding covariates increases the number of parameters
386: : 12 for <strong>model=V1, </strong>16 for <strong>model=V1+V1*age
387: </strong>and 20 for <strong>model=V1+V2+V3.</strong></p>
388:
1.6 lievre 389: <h4><font color="#FF0000">Guess values for optimization</font><font
390: color="#00006A"> </font></h4>
391:
392: <p>You must write the initial guess values of the parameters for
393: optimization. The number of parameters, <em>N</em> depends on the
1.2 lievre 394: number of absorbing states and non-absorbing states and on the
395: number of covariates. <br>
396: <em>N</em> is given by the formula <em>N</em>=(<em>nlstate</em> +
397: <em>ndeath</em>-1)*<em>nlstate</em>*<em>ncov</em> . <br>
398: <br>
399: Thus in the simple case with 2 covariates (the model is log
400: (pij/pii) = aij + bij * age where intercept and age are the two
401: covariates), and 2 health degrees (1 for disability-free and 2
402: for disability) and 1 absorbing state (3), you must enter 8
403: initials values, a12, b12, a13, b13, a21, b21, a23, b23. You can
404: start with zeros as in this example, but if you have a more
405: precise set (for example from an earlier run) you can enter it
406: and it will speed up them<br>
1.4 lievre 407: Each of the four lines starts with indices "ij": <b>ij
1.6 lievre 408: aij bij</b> </p>
409:
410: <blockquote>
411: <pre># Guess values of aij and bij in log (pij/pii) = aij + bij * age
412: 12 -14.155633 0.110794
413: 13 -7.925360 0.032091
414: 21 -1.890135 -0.029473
415: 23 -6.234642 0.022315 </pre>
416: </blockquote>
417:
1.8 lievre 418: <p>or, to simplify (in most of cases it converges but there is no
419: warranty!): </p>
1.6 lievre 420:
421: <blockquote>
422: <pre>12 0.0 0.0
423: 13 0.0 0.0
424: 21 0.0 0.0
425: 23 0.0 0.0</pre>
426: </blockquote>
427:
1.9 ! brouard 428: <p> In order to speed up the convergence you can make a first run with
! 429: a large stepm i.e stepm=12 or 24 and then decrease the stepm until
! 430: stepm=1 month. If newstepm is the new shorter stepm and stepm can be
! 431: expressed as a multiple of newstepm, like newstepm=n stepm, then the
! 432: following approximation holds:
! 433: <pre>aij(n stepm) = aij(stepm) +ln(n)
! 434: </pre> and
! 435: <pre>bij(n stepm) = bij(stepm) .</pre>
1.6 lievre 436: <h4><font color="#FF0000">Guess values for computing variances</font></h4>
1.2 lievre 437:
1.6 lievre 438: <p>This is an output if <a href="#mle">mle</a>=1. But it can be
439: used as an input to get the various output data files (Health
1.2 lievre 440: expectancies, stationary prevalence etc.) and figures without
1.6 lievre 441: rerunning the rather long maximisation phase (mle=0). </p>
1.2 lievre 442:
1.6 lievre 443: <p>The scales are small values for the evaluation of numerical
1.2 lievre 444: derivatives. These derivatives are used to compute the hessian
445: matrix of the parameters, that is the inverse of the covariance
446: matrix, and the variances of health expectancies. Each line
447: consists in indices "ij" followed by the initial scales
1.6 lievre 448: (zero to simplify) associated with aij and bij. </p>
1.2 lievre 449:
1.6 lievre 450: <ul>
451: <li>If mle=1 you can enter zeros:</li>
1.2 lievre 452: </ul>
453:
1.6 lievre 454: <blockquote>
455: <pre># Scales (for hessian or gradient estimation)
456: 12 0. 0.
457: 13 0. 0.
458: 21 0. 0.
459: 23 0. 0. </pre>
460: </blockquote>
461:
462: <ul>
463: <li>If mle=0 you must enter a covariance matrix (usually
464: obtained from an earlier run).</li>
1.2 lievre 465: </ul>
466:
1.6 lievre 467: <h4><font color="#FF0000">Covariance matrix of parameters</font></h4>
468:
469: <p>This is an output if <a href="#mle">mle</a>=1. But it can be
470: used as an input to get the various output data files (Health
1.5 lievre 471: expectancies, stationary prevalence etc.) and figures without
1.6 lievre 472: rerunning the rather long maximisation phase (mle=0). </p>
1.2 lievre 473:
1.6 lievre 474: <p>Each line starts with indices "ijk" followed by the
475: covariances between aij and bij: </p>
1.5 lievre 476:
1.6 lievre 477: <pre>
478: 121 Var(a12)
479: 122 Cov(b12,a12) Var(b12)
480: ...
481: 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23) </pre>
1.2 lievre 482:
1.6 lievre 483: <ul>
484: <li>If mle=1 you can enter zeros. </li>
1.2 lievre 485: </ul>
486:
1.6 lievre 487: <blockquote>
488: <pre># Covariance matrix
489: 121 0.
490: 122 0. 0.
491: 131 0. 0. 0.
492: 132 0. 0. 0. 0.
493: 211 0. 0. 0. 0. 0.
494: 212 0. 0. 0. 0. 0. 0.
495: 231 0. 0. 0. 0. 0. 0. 0.
496: 232 0. 0. 0. 0. 0. 0. 0. 0.</pre>
497: </blockquote>
498:
499: <ul>
500: <li>If mle=0 you must enter a covariance matrix (usually
501: obtained from an earlier run).<br>
502: </li>
1.2 lievre 503: </ul>
504:
1.6 lievre 505: <h4><font color="#FF0000">Age range for calculation of stationary
506: prevalences and health expectancies</font></h4>
1.2 lievre 507:
1.6 lievre 508: <pre>agemin=70 agemax=100 bage=50 fage=100</pre>
1.2 lievre 509:
1.6 lievre 510: <p>Once we obtained the estimated parameters, the program is able
511: to calculated stationary prevalence, transitions probabilities
512: and life expectancies at any age. Choice of age range is useful
513: for extrapolation. In our data file, ages varies from age 70 to
1.8 lievre 514: 102. It is possible to get extrapolated stationary prevalence by
515: age ranging from agemin to agemax. </p>
1.6 lievre 516:
1.8 lievre 517: <p>Setting bage=50 (begin age) and fage=100 (final age), makes
518: the program computing life expectancy from age 'bage' to age
519: 'fage'. As we use a model, we can interessingly compute life
520: expectancy on a wider age range than the age range from the data.
521: But the model can be rather wrong on much larger intervals.
522: Program is limited to around 120 for upper age!</p>
1.6 lievre 523:
524: <ul>
525: <li><b>agemin=</b> Minimum age for calculation of the
526: stationary prevalence </li>
527: <li><b>agemax=</b> Maximum age for calculation of the
528: stationary prevalence </li>
529: <li><b>bage=</b> Minimum age for calculation of the health
530: expectancies </li>
531: <li><b>fage=</b> Maximum age for calculation of the health
532: expectancies </li>
1.2 lievre 533: </ul>
534:
1.6 lievre 535: <h4><a name="Computing"><font color="#FF0000">Computing</font></a><font
536: color="#FF0000"> the observed prevalence</font></h4>
1.4 lievre 537:
1.6 lievre 538: <pre>begin-prev-date=1/1/1984 end-prev-date=1/6/1988 </pre>
1.4 lievre 539:
1.6 lievre 540: <p>Statements 'begin-prev-date' and 'end-prev-date' allow to
541: select the period in which we calculate the observed prevalences
542: in each state. In this example, the prevalences are calculated on
543: data survey collected between 1 january 1984 and 1 june 1988. </p>
544:
545: <ul>
546: <li><strong>begin-prev-date= </strong>Starting date
547: (day/month/year)</li>
548: <li><strong>end-prev-date= </strong>Final date
549: (day/month/year)</li>
1.4 lievre 550: </ul>
551:
1.6 lievre 552: <h4><font color="#FF0000">Population- or status-based health
553: expectancies</font></h4>
1.5 lievre 554:
1.6 lievre 555: <pre>pop_based=0</pre>
1.5 lievre 556:
1.8 lievre 557: <p>The program computes status-based health expectancies, i.e
558: health expectancies which depends on your initial health state.
559: If you are healthy your healthy life expectancy (e11) is higher
560: than if you were disabled (e21, with e11 > e21).<br>
561: To compute a healthy life expectancy independant of the initial
562: status we have to weight e11 and e21 according to the probability
563: to be in each state at initial age or, with other word, according
564: to the proportion of people in each state.<br>
565: We prefer computing a 'pure' period healthy life expectancy based
566: only on the transtion forces. Then the weights are simply the
567: stationnary prevalences or 'implied' prevalences at the initial
568: age.<br>
569: Some other people would like to use the cross-sectional
570: prevalences (the "Sullivan prevalences") observed at
571: the initial age during a period of time <a href="#Computing">defined
572: just above</a>. </p>
1.7 brouard 573:
574: <ul>
1.8 lievre 575: <li><strong>popbased= 0 </strong>Health expectancies are
576: computed at each age from stationary prevalences
577: 'expected' at this initial age.</li>
578: <li><strong>popbased= 1 </strong>Health expectancies are
579: computed at each age from cross-sectional 'observed'
580: prevalence at this initial age. As all the population is
581: not observed at the same exact date we define a short
582: period were the observed prevalence is computed.</li>
1.7 brouard 583: </ul>
584:
585: <h4><font color="#FF0000">Prevalence forecasting ( Experimental)</font></h4>
1.5 lievre 586:
1.6 lievre 587: <pre>starting-proj-date=1/1/1989 final-proj-date=1/1/1992 mov_average=0 </pre>
1.5 lievre 588:
1.8 lievre 589: <p>Prevalence and population projections are only available if
590: the interpolation unit is a month, i.e. stepm=1 and if there are
591: no covariate. The programme estimates the prevalence in each
592: state at a precise date expressed in day/month/year. The
593: programme computes one forecasted prevalence a year from a
594: starting date (1 january of 1989 in this example) to a final date
595: (1 january 1992). The statement mov_average allows to compute
596: smoothed forecasted prevalences with a five-age moving average
597: centered at the mid-age of the five-age period. </p>
1.5 lievre 598:
1.6 lievre 599: <ul>
600: <li><strong>starting-proj-date</strong>= starting date
601: (day/month/year) of forecasting</li>
602: <li><strong>final-proj-date= </strong>final date
603: (day/month/year) of forecasting</li>
604: <li><strong>mov_average</strong>= smoothing with a five-age
605: moving average centered at the mid-age of the five-age
606: period. The command<strong> mov_average</strong> takes
607: value 1 if the prevalences are smoothed and 0 otherwise.</li>
608: </ul>
1.5 lievre 609:
1.6 lievre 610: <h4><font color="#FF0000">Last uncommented line : Population
611: forecasting </font></h4>
1.5 lievre 612:
1.6 lievre 613: <pre>popforecast=0 popfile=pyram.txt popfiledate=1/1/1989 last-popfiledate=1/1/1992</pre>
1.5 lievre 614:
1.6 lievre 615: <p>This command is available if the interpolation unit is a
1.7 brouard 616: month, i.e. stepm=1 and if popforecast=1. From a data file
617: including age and number of persons alive at the precise date
618: ‘popfiledate’, you can forecast the number of persons
619: in each state until date ‘last-popfiledate’. In this
620: example, the popfile <a href="pyram.txt"><b>pyram.txt</b></a>
621: includes real data which are the Japanese population in 1989.</p>
622:
623: <ul type="disc">
624: <li class="MsoNormal"
625: style="TEXT-ALIGN: justify; mso-margin-top-alt: auto; mso-margin-bottom-alt: auto; mso-list: l10 level1 lfo36; tab-stops: list 36.0pt"><b>popforecast=
626: 0 </b>Option for population forecasting. If
627: popforecast=1, the programme does the forecasting<b>.</b></li>
628: <li class="MsoNormal"
629: style="TEXT-ALIGN: justify; mso-margin-top-alt: auto; mso-margin-bottom-alt: auto; mso-list: l10 level1 lfo36; tab-stops: list 36.0pt"><b>popfile=
630: </b>name of the population file</li>
631: <li class="MsoNormal"
632: style="TEXT-ALIGN: justify; mso-margin-top-alt: auto; mso-margin-bottom-alt: auto; mso-list: l10 level1 lfo36; tab-stops: list 36.0pt"><b>popfiledate=</b>
633: date of the population population</li>
634: <li class="MsoNormal"
635: style="TEXT-ALIGN: justify; mso-margin-top-alt: auto; mso-margin-bottom-alt: auto; mso-list: l10 level1 lfo36; tab-stops: list 36.0pt"><b>last-popfiledate</b>=
636: date of the last population projection </li>
637: </ul>
1.5 lievre 638:
1.6 lievre 639: <hr>
1.5 lievre 640:
1.6 lievre 641: <h2><a name="running"></a><font color="#00006A">Running Imach
642: with this example</font></h2>
1.5 lievre 643:
1.6 lievre 644: <p>We assume that you entered your <a href="biaspar.imach">1st_example
645: parameter file</a> as explained <a href="#biaspar">above</a>. To
646: run the program you should click on the imach.exe icon and enter
647: the name of the parameter file which is for example <a
648: href="C:\usr\imach\mle\biaspar.txt">C:\usr\imach\mle\biaspar.txt</a>
649: (you also can click on the biaspar.txt icon located in <br>
650: <a href="C:\usr\imach\mle">C:\usr\imach\mle</a> and put it with
651: the mouse on the imach window).<br>
652: </p>
653:
654: <p>The time to converge depends on the step unit that you used (1
655: month is cpu consuming), on the number of cases, and on the
656: number of variables.</p>
1.5 lievre 657:
1.6 lievre 658: <p>The program outputs many files. Most of them are files which
659: will be plotted for better understanding.</p>
1.5 lievre 660:
1.6 lievre 661: <hr>
1.5 lievre 662:
1.6 lievre 663: <h2><a name="output"><font color="#00006A">Output of the program
664: and graphs</font> </a></h2>
1.5 lievre 665:
1.6 lievre 666: <p>Once the optimization is finished, some graphics can be made
667: with a grapher. We use Gnuplot which is an interactive plotting
668: program copyrighted but freely distributed. A gnuplot reference
669: manual is available <a href="http://www.gnuplot.info/">here</a>. <br>
670: When the running is finished, the user should enter a caracter
671: for plotting and output editing. </p>
672:
673: <p>These caracters are:</p>
674:
675: <ul>
676: <li>'c' to start again the program from the beginning.</li>
677: <li>'e' opens the <a href="biaspar.htm"><strong>biaspar.htm</strong></a>
678: file to edit the output files and graphs. </li>
679: <li>'q' for exiting.</li>
680: </ul>
1.4 lievre 681:
1.6 lievre 682: <h5><font size="4"><strong>Results files </strong></font><br>
683: <br>
684: <font color="#EC5E5E" size="3"><strong>- </strong></font><a
685: name="Observed prevalence in each state"><font color="#EC5E5E"
686: size="3"><strong>Observed prevalence in each state</strong></font></a><font
687: color="#EC5E5E" size="3"><strong> (and at first pass)</strong></font><b>:
688: </b><a href="prbiaspar.txt"><b>prbiaspar.txt</b></a><br>
689: </h5>
690:
691: <p>The first line is the title and displays each field of the
692: file. The first column is age. The fields 2 and 6 are the
693: proportion of individuals in states 1 and 2 respectively as
694: observed during the first exam. Others fields are the numbers of
695: people in states 1, 2 or more. The number of columns increases if
696: the number of states is higher than 2.<br>
697: The header of the file is </p>
698:
699: <pre># Age Prev(1) N(1) N Age Prev(2) N(2) N
700: 70 1.00000 631 631 70 0.00000 0 631
701: 71 0.99681 625 627 71 0.00319 2 627
702: 72 0.97125 1115 1148 72 0.02875 33 1148 </pre>
703:
704: <p>It means that at age 70, the prevalence in state 1 is 1.000
705: and in state 2 is 0.00 . At age 71 the number of individuals in
706: state 1 is 625 and in state 2 is 2, hence the total number of
707: people aged 71 is 625+2=627. <br>
708: </p>
709:
710: <h5><font color="#EC5E5E" size="3"><b>- Estimated parameters and
711: covariance matrix</b></font><b>: </b><a href="rbiaspar.txt"><b>rbiaspar.txt</b></a></h5>
712:
713: <p>This file contains all the maximisation results: </p>
714:
715: <pre> -2 log likelihood= 21660.918613445392
716: Estimated parameters: a12 = -12.290174 b12 = 0.092161
717: a13 = -9.155590 b13 = 0.046627
718: a21 = -2.629849 b21 = -0.022030
719: a23 = -7.958519 b23 = 0.042614
720: Covariance matrix: Var(a12) = 1.47453e-001
721: Var(b12) = 2.18676e-005
722: Var(a13) = 2.09715e-001
723: Var(b13) = 3.28937e-005
724: Var(a21) = 9.19832e-001
725: Var(b21) = 1.29229e-004
726: Var(a23) = 4.48405e-001
727: Var(b23) = 5.85631e-005
728: </pre>
729:
730: <p>By substitution of these parameters in the regression model,
731: we obtain the elementary transition probabilities:</p>
732:
733: <p><img src="pebiaspar1.gif" width="400" height="300"></p>
734:
735: <h5><font color="#EC5E5E" size="3"><b>- Transition probabilities</b></font><b>:
736: </b><a href="pijrbiaspar.txt"><b>pijrbiaspar.txt</b></a></h5>
737:
738: <p>Here are the transitions probabilities Pij(x, x+nh) where nh
739: is a multiple of 2 years. The first column is the starting age x
740: (from age 50 to 100), the second is age (x+nh) and the others are
741: the transition probabilities p11, p12, p13, p21, p22, p23. For
742: example, line 5 of the file is: </p>
743:
744: <pre> 100 106 0.02655 0.17622 0.79722 0.01809 0.13678 0.84513 </pre>
745:
746: <p>and this means: </p>
747:
748: <pre>p11(100,106)=0.02655
749: p12(100,106)=0.17622
750: p13(100,106)=0.79722
751: p21(100,106)=0.01809
752: p22(100,106)=0.13678
753: p22(100,106)=0.84513 </pre>
754:
755: <h5><font color="#EC5E5E" size="3"><b>- </b></font><a
756: name="Stationary prevalence in each state"><font color="#EC5E5E"
757: size="3"><b>Stationary prevalence in each state</b></font></a><b>:
758: </b><a href="plrbiaspar.txt"><b>plrbiaspar.txt</b></a></h5>
759:
760: <pre>#Prevalence
761: #Age 1-1 2-2
762:
763: #************
764: 70 0.90134 0.09866
765: 71 0.89177 0.10823
766: 72 0.88139 0.11861
767: 73 0.87015 0.12985 </pre>
1.4 lievre 768:
1.6 lievre 769: <p>At age 70 the stationary prevalence is 0.90134 in state 1 and
1.3 lievre 770: 0.09866 in state 2. This stationary prevalence differs from
1.2 lievre 771: observed prevalence. Here is the point. The observed prevalence
772: at age 70 results from the incidence of disability, incidence of
773: recovery and mortality which occurred in the past of the cohort.
774: Stationary prevalence results from a simulation with actual
775: incidences and mortality (estimated from this cross-longitudinal
776: survey). It is the best predictive value of the prevalence in the
777: future if "nothing changes in the future". This is
778: exactly what demographers do with a Life table. Life expectancy
779: is the expected mean time to survive if observed mortality rates
780: (incidence of mortality) "remains constant" in the
1.6 lievre 781: future. </p>
1.2 lievre 782:
1.6 lievre 783: <h5><font color="#EC5E5E" size="3"><b>- Standard deviation of
784: stationary prevalence</b></font><b>: </b><a
785: href="vplrbiaspar.txt"><b>vplrbiaspar.txt</b></a></h5>
786:
787: <p>The stationary prevalence has to be compared with the observed
1.2 lievre 788: prevalence by age. But both are statistical estimates and
789: subjected to stochastic errors due to the size of the sample, the
790: design of the survey, and, for the stationary prevalence to the
791: model used and fitted. It is possible to compute the standard
1.6 lievre 792: deviation of the stationary prevalence at each age.</p>
793:
794: <h5><font color="#EC5E5E" size="3">-Observed and stationary
795: prevalence in state (2=disable) with the confident interval</font>:<b>
796: </b><a href="vbiaspar21.htm"><b>vbiaspar21.gif</b></a></h5>
797:
798: <p>This graph exhibits the stationary prevalence in state (2)
799: with the confidence interval in red. The green curve is the
800: observed prevalence (or proportion of individuals in state (2)).
801: Without discussing the results (it is not the purpose here), we
802: observe that the green curve is rather below the stationary
803: prevalence. It suggests an increase of the disability prevalence
804: in the future.</p>
805:
806: <p><img src="vbiaspar21.gif" width="400" height="300"></p>
807:
808: <h5><font color="#EC5E5E" size="3"><b>-Convergence to the
809: stationary prevalence of disability</b></font><b>: </b><a
810: href="pbiaspar11.gif"><b>pbiaspar11.gif</b></a><br>
811: <img src="pbiaspar11.gif" width="400" height="300"> </h5>
1.2 lievre 812:
1.6 lievre 813: <p>This graph plots the conditional transition probabilities from
814: an initial state (1=healthy in red at the bottom, or 2=disable in
1.2 lievre 815: green on top) at age <em>x </em>to the final state 2=disable<em> </em>at
816: age <em>x+h. </em>Conditional means at the condition to be alive
817: at age <em>x+h </em>which is <i>hP12x</i> + <em>hP22x</em>. The
818: curves <i>hP12x/(hP12x</i> + <em>hP22x) </em>and <i>hP22x/(hP12x</i>
819: + <em>hP22x) </em>converge with <em>h, </em>to the <em>stationary
820: prevalence of disability</em>. In order to get the stationary
821: prevalence at age 70 we should start the process at an earlier
822: age, i.e.50. If the disability state is defined by severe
823: disability criteria with only a few chance to recover, then the
824: incidence of recovery is low and the time to convergence is
1.6 lievre 825: probably longer. But we don't have experience yet.</p>
1.2 lievre 826:
1.6 lievre 827: <h5><font color="#EC5E5E" size="3"><b>- Life expectancies by age
828: and initial health status</b></font><b>: </b><a
829: href="erbiaspar.txt"><b>erbiaspar.txt</b></a></h5>
830:
831: <pre># Health expectancies
832: # Age 1-1 1-2 2-1 2-2
833: 70 10.9226 3.0401 5.6488 6.2122
834: 71 10.4384 3.0461 5.2477 6.1599
835: 72 9.9667 3.0502 4.8663 6.1025
836: 73 9.5077 3.0524 4.5044 6.0401 </pre>
1.2 lievre 837:
1.7 brouard 838: <pre>For example 70 10.4227 3.0402 5.6488 5.7123 means:
839: e11=10.4227 e12=3.0402 e21=5.6488 e22=5.7123</pre>
1.2 lievre 840:
1.6 lievre 841: <pre><img src="expbiaspar21.gif" width="400" height="300"><img
842: src="expbiaspar11.gif" width="400" height="300"></pre>
1.2 lievre 843:
1.6 lievre 844: <p>For example, life expectancy of a healthy individual at age 70
1.7 brouard 845: is 10.42 in the healthy state and 3.04 in the disability state
846: (=13.46 years). If he was disable at age 70, his life expectancy
847: will be shorter, 5.64 in the healthy state and 5.71 in the
848: disability state (=11.35 years). The total life expectancy is a
849: weighted mean of both, 13.46 and 11.35; weight is the proportion
1.2 lievre 850: of people disabled at age 70. In order to get a pure period index
1.6 lievre 851: (i.e. based only on incidences) we use the <a
852: href="#Stationary prevalence in each state">computed or
853: stationary prevalence</a> at age 70 (i.e. computed from
854: incidences at earlier ages) instead of the <a
855: href="#Observed prevalence in each state">observed prevalence</a>
856: (for example at first exam) (<a href="#Health expectancies">see
857: below</a>).</p>
858:
859: <h5><font color="#EC5E5E" size="3"><b>- Variances of life
860: expectancies by age and initial health status</b></font><b>: </b><a
861: href="vrbiaspar.txt"><b>vrbiaspar.txt</b></a></h5>
862:
863: <p>For example, the covariances of life expectancies Cov(ei,ej)
864: at age 50 are (line 3) </p>
865:
866: <pre> Cov(e1,e1)=0.4776 Cov(e1,e2)=0.0488=Cov(e2,e1) Cov(e2,e2)=0.0424</pre>
867:
868: <h5><font color="#EC5E5E" size="3"><b>- </b></font><a
869: name="Health expectancies"><font color="#EC5E5E" size="3"><b>Health
870: expectancies</b></font></a><font color="#EC5E5E" size="3"><b>
871: with standard errors in parentheses</b></font><b>: </b><a
872: href="trbiaspar.txt"><font face="Courier New"><b>trbiaspar.txt</b></font></a></h5>
873:
874: <pre>#Total LEs with variances: e.. (std) e.1 (std) e.2 (std) </pre>
875:
1.7 brouard 876: <pre>70 13.26 (0.22) 9.95 (0.20) 3.30 (0.14) </pre>
1.6 lievre 877:
1.7 brouard 878: <p>Thus, at age 70 the total life expectancy, e..=13.26 years is
879: the weighted mean of e1.=13.46 and e2.=11.35 by the stationary
1.3 lievre 880: prevalence at age 70 which are 0.90134 in state 1 and 0.09866 in
1.7 brouard 881: state 2, respectively (the sum is equal to one). e.1=9.95 is the
1.2 lievre 882: Disability-free life expectancy at age 70 (it is again a weighted
1.7 brouard 883: mean of e11 and e21). e.2=3.30 is also the life expectancy at age
1.6 lievre 884: 70 to be spent in the disability state.</p>
1.2 lievre 885:
1.6 lievre 886: <h5><font color="#EC5E5E" size="3"><b>-Total life expectancy by
887: age and health expectancies in states (1=healthy) and (2=disable)</b></font><b>:
888: </b><a href="ebiaspar1.gif"><b>ebiaspar1.gif</b></a></h5>
889:
890: <p>This figure represents the health expectancies and the total
891: life expectancy with the confident interval in dashed curve. </p>
892:
893: <pre> <img src="ebiaspar1.gif" width="400" height="300"></pre>
894:
895: <p>Standard deviations (obtained from the information matrix of
896: the model) of these quantities are very useful.
897: Cross-longitudinal surveys are costly and do not involve huge
898: samples, generally a few thousands; therefore it is very
899: important to have an idea of the standard deviation of our
900: estimates. It has been a big challenge to compute the Health
901: Expectancy standard deviations. Don't be confuse: life expectancy
902: is, as any expected value, the mean of a distribution; but here
903: we are not computing the standard deviation of the distribution,
904: but the standard deviation of the estimate of the mean.</p>
905:
906: <p>Our health expectancies estimates vary according to the sample
907: size (and the standard deviations give confidence intervals of
908: the estimate) but also according to the model fitted. Let us
909: explain it in more details.</p>
910:
911: <p>Choosing a model means ar least two kind of choices. First we
912: have to decide the number of disability states. Second we have to
913: design, within the logit model family, the model: variables,
914: covariables, confonding factors etc. to be included.</p>
915:
916: <p>More disability states we have, better is our demographical
917: approach of the disability process, but smaller are the number of
1.2 lievre 918: transitions between each state and higher is the noise in the
919: measurement. We do not have enough experiments of the various
920: models to summarize the advantages and disadvantages, but it is
921: important to say that even if we had huge and unbiased samples,
922: the total life expectancy computed from a cross-longitudinal
923: survey, varies with the number of states. If we define only two
924: states, alive or dead, we find the usual life expectancy where it
925: is assumed that at each age, people are at the same risk to die.
926: If we are differentiating the alive state into healthy and
927: disable, and as the mortality from the disability state is higher
928: than the mortality from the healthy state, we are introducing
929: heterogeneity in the risk of dying. The total mortality at each
930: age is the weighted mean of the mortality in each state by the
931: prevalence in each state. Therefore if the proportion of people
932: at each age and in each state is different from the stationary
933: equilibrium, there is no reason to find the same total mortality
934: at a particular age. Life expectancy, even if it is a very useful
935: tool, has a very strong hypothesis of homogeneity of the
936: population. Our main purpose is not to measure differential
937: mortality but to measure the expected time in a healthy or
938: disability state in order to maximise the former and minimize the
939: latter. But the differential in mortality complexifies the
1.6 lievre 940: measurement.</p>
1.2 lievre 941:
1.6 lievre 942: <p>Incidences of disability or recovery are not affected by the
943: number of states if these states are independant. But incidences
944: estimates are dependant on the specification of the model. More
945: covariates we added in the logit model better is the model, but
946: some covariates are not well measured, some are confounding
947: factors like in any statistical model. The procedure to "fit
948: the best model' is similar to logistic regression which itself is
949: similar to regression analysis. We haven't yet been sofar because
950: we also have a severe limitation which is the speed of the
951: convergence. On a Pentium III, 500 MHz, even the simplest model,
952: estimated by month on 8,000 people may take 4 hours to converge.
953: Also, the program is not yet a statistical package, which permits
954: a simple writing of the variables and the model to take into
955: account in the maximisation. The actual program allows only to
956: add simple variables like age+sex or age+sex+ age*sex but will
957: never be general enough. But what is to remember, is that
1.2 lievre 958: incidences or probability of change from one state to another is
1.6 lievre 959: affected by the variables specified into the model.</p>
1.2 lievre 960:
1.6 lievre 961: <p>Also, the age range of the people interviewed has a link with
962: the age range of the life expectancy which can be estimated by
1.2 lievre 963: extrapolation. If your sample ranges from age 70 to 95, you can
964: clearly estimate a life expectancy at age 70 and trust your
965: confidence interval which is mostly based on your sample size,
966: but if you want to estimate the life expectancy at age 50, you
967: should rely in your model, but fitting a logistic model on a age
1.6 lievre 968: range of 70-95 and estimating probabilties of transition out of
1.2 lievre 969: this age range, say at age 50 is very dangerous. At least you
970: should remember that the confidence interval given by the
971: standard deviation of the health expectancies, are under the
972: strong assumption that your model is the 'true model', which is
1.6 lievre 973: probably not the case.</p>
1.5 lievre 974:
1.6 lievre 975: <h5><font color="#EC5E5E" size="3"><b>- Copy of the parameter
976: file</b></font><b>: </b><a href="orbiaspar.txt"><b>orbiaspar.txt</b></a></h5>
1.2 lievre 977:
1.6 lievre 978: <p>This copy of the parameter file can be useful to re-run the
979: program while saving the old output files. </p>
1.2 lievre 980:
1.6 lievre 981: <h5><font color="#EC5E5E" size="3"><b>- Prevalence forecasting</b></font><b>:
982: </b><a href="frbiaspar.txt"><b>frbiaspar.txt</b></a></h5>
1.2 lievre 983:
1.7 brouard 984: <p
985: style="TEXT-ALIGN: justify; tab-stops: 45.8pt 91.6pt 137.4pt 183.2pt 229.0pt 274.8pt 320.6pt 366.4pt 412.2pt 458.0pt 503.8pt 549.6pt 595.4pt 641.2pt 687.0pt 732.8pt">First,
986: we have estimated the observed prevalence between 1/1/1984 and
987: 1/6/1988. The mean date of interview (weighed average of the
988: interviews performed between1/1/1984 and 1/6/1988) is estimated
989: to be 13/9/1985, as written on the top on the file. Then we
990: forecast the probability to be in each state. </p>
991:
992: <p
993: style="TEXT-ALIGN: justify; tab-stops: 45.8pt 91.6pt 137.4pt 183.2pt 229.0pt 274.8pt 320.6pt 366.4pt 412.2pt 458.0pt 503.8pt 549.6pt 595.4pt 641.2pt 687.0pt 732.8pt">Example,
994: at date 1/1/1989 : </p>
995:
996: <pre class="MsoNormal"># StartingAge FinalAge P.1 P.2 P.3
997: # Forecasting at date 1/1/1989
998: 73 0.807 0.078 0.115</pre>
999:
1000: <p
1001: style="TEXT-ALIGN: justify; tab-stops: 45.8pt 91.6pt 137.4pt 183.2pt 229.0pt 274.8pt 320.6pt 366.4pt 412.2pt 458.0pt 503.8pt 549.6pt 595.4pt 641.2pt 687.0pt 732.8pt">Since
1002: the minimum age is 70 on the 13/9/1985, the youngest forecasted
1003: age is 73. This means that at age a person aged 70 at 13/9/1989
1004: has a probability to enter state1 of 0.807 at age 73 on 1/1/1989.
1005: Similarly, the probability to be in state 2 is 0.078 and the
1006: probability to die is 0.115. Then, on the 1/1/1989, the
1007: prevalence of disability at age 73 is estimated to be 0.088.</p>
1.4 lievre 1008:
1.6 lievre 1009: <h5><font color="#EC5E5E" size="3"><b>- Population forecasting</b></font><b>:
1010: </b><a href="poprbiaspar.txt"><b>poprbiaspar.txt</b></a></h5>
1.4 lievre 1011:
1.6 lievre 1012: <pre># Age P.1 P.2 P.3 [Population]
1013: # Forecasting at date 1/1/1989
1014: 75 572685.22 83798.08
1015: 74 621296.51 79767.99
1016: 73 645857.70 69320.60 </pre>
1.4 lievre 1017:
1.6 lievre 1018: <pre># Forecasting at date 1/1/19909
1019: 76 442986.68 92721.14 120775.48
1020: 75 487781.02 91367.97 121915.51
1021: 74 512892.07 85003.47 117282.76 </pre>
1.4 lievre 1022:
1.7 brouard 1023: <p>From the population file, we estimate the number of people in
1024: each state. At age 73, 645857 persons are in state 1 and 69320
1025: are in state 2. One year latter, 512892 are still in state 1,
1026: 85003 are in state 2 and 117282 died before 1/1/1990.</p>
1027:
1.6 lievre 1028: <hr>
1.4 lievre 1029:
1.8 lievre 1030: <h2><a name="example"></a><font color="#00006A">Trying an example</font></h2>
1.5 lievre 1031:
1.6 lievre 1032: <p>Since you know how to run the program, it is time to test it
1033: on your own computer. Try for example on a parameter file named <a
1034: href="..\mytry\imachpar.txt">imachpar.txt</a> which is a copy of <font
1035: size="2" face="Courier New">mypar.txt</font> included in the
1036: subdirectory of imach, <font size="2" face="Courier New">mytry</font>.
1037: Edit it to change the name of the data file to <font size="2"
1038: face="Courier New">..\data\mydata.txt</font> if you don't want to
1039: copy it on the same directory. The file <font face="Courier New">mydata.txt</font>
1040: is a smaller file of 3,000 people but still with 4 waves. </p>
1.5 lievre 1041:
1.6 lievre 1042: <p>Click on the imach.exe icon to open a window. Answer to the
1043: question:'<strong>Enter the parameter file name:'</strong></p>
1.5 lievre 1044:
1.6 lievre 1045: <table border="1">
1.2 lievre 1046: <tr>
1.7 brouard 1047: <td width="100%"><strong>IMACH, Version 0.71</strong><p><strong>Enter
1.6 lievre 1048: the parameter file name: ..\mytry\imachpar.txt</strong></p>
1.2 lievre 1049: </td>
1050: </tr>
1051: </table>
1052:
1.6 lievre 1053: <p>Most of the data files or image files generated, will use the
1.2 lievre 1054: 'imachpar' string into their name. The running time is about 2-3
1055: minutes on a Pentium III. If the execution worked correctly, the
1056: outputs files are created in the current directory, and should be
1.6 lievre 1057: the same as the mypar files initially included in the directory <font
1058: size="2" face="Courier New">mytry</font>.</p>
1.5 lievre 1059:
1.6 lievre 1060: <ul>
1061: <li><pre><u>Output on the screen</u> The output screen looks like <a
1062: href="imachrun.LOG">this Log file</a>
1063: #
1.5 lievre 1064:
1.6 lievre 1065: title=MLE datafile=..\data\mydata.txt lastobs=3000 firstpass=1 lastpass=3
1066: ftol=1.000000e-008 stepm=24 ncov=2 nlstate=2 ndeath=1 maxwav=4 mle=1 weight=0</pre>
1067: </li>
1068: <li><pre>Total number of individuals= 2965, Agemin = 70.00, Agemax= 100.92
1.5 lievre 1069:
1.6 lievre 1070: Warning, no any valid information for:126 line=126
1071: Warning, no any valid information for:2307 line=2307
1072: Delay (in months) between two waves Min=21 Max=51 Mean=24.495826
1073: <font face="Times New Roman">These lines give some warnings on the data file and also some raw statistics on frequencies of transitions.</font>
1074: Age 70 1.=230 loss[1]=3.5% 2.=16 loss[2]=12.5% 1.=222 prev[1]=94.1% 2.=14
1075: prev[2]=5.9% 1-1=8 11=200 12=7 13=15 2-1=2 21=6 22=7 23=1
1076: Age 102 1.=0 loss[1]=NaNQ% 2.=0 loss[2]=NaNQ% 1.=0 prev[1]=NaNQ% 2.=0 </pre>
1077: </li>
1078: </ul>
1.2 lievre 1079:
1.6 lievre 1080: <p> </p>
1.2 lievre 1081:
1.6 lievre 1082: <ul>
1083: <li>Maximisation with the Powell algorithm. 8 directions are
1084: given corresponding to the 8 parameters. this can be
1085: rather long to get convergence.<br>
1086: <font size="1" face="Courier New"><br>
1.2 lievre 1087: Powell iter=1 -2*LL=11531.405658264877 1 0.000000000000 2
1088: 0.000000000000 3<br>
1089: 0.000000000000 4 0.000000000000 5 0.000000000000 6
1090: 0.000000000000 7 <br>
1091: 0.000000000000 8 0.000000000000<br>
1092: 1..........2.................3..........4.................5.........<br>
1093: 6................7........8...............<br>
1094: Powell iter=23 -2*LL=6744.954108371555 1 -12.967632334283
1095: <br>
1096: 2 0.135136681033 3 -7.402109728262 4 0.067844593326 <br>
1097: 5 -0.673601538129 6 -0.006615504377 7 -5.051341616718 <br>
1098: 8 0.051272038506<br>
1099: 1..............2...........3..............4...........<br>
1100: 5..........6................7...........8.........<br>
1101: #Number of iterations = 23, -2 Log likelihood =
1102: 6744.954042573691<br>
1103: # Parameters<br>
1104: 12 -12.966061 0.135117 <br>
1105: 13 -7.401109 0.067831 <br>
1106: 21 -0.672648 -0.006627 <br>
1.6 lievre 1107: 23 -5.051297 0.051271 </font><br>
1108: </li>
1109: <li><pre><font size="2">Calculation of the hessian matrix. Wait...
1110: 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
1111:
1112: Inverting the hessian to get the covariance matrix. Wait...
1113:
1114: #Hessian matrix#
1115: 3.344e+002 2.708e+004 -4.586e+001 -3.806e+003 -1.577e+000 -1.313e+002 3.914e-001 3.166e+001
1116: 2.708e+004 2.204e+006 -3.805e+003 -3.174e+005 -1.303e+002 -1.091e+004 2.967e+001 2.399e+003
1117: -4.586e+001 -3.805e+003 4.044e+002 3.197e+004 2.431e-002 1.995e+000 1.783e-001 1.486e+001
1118: -3.806e+003 -3.174e+005 3.197e+004 2.541e+006 2.436e+000 2.051e+002 1.483e+001 1.244e+003
1119: -1.577e+000 -1.303e+002 2.431e-002 2.436e+000 1.093e+002 8.979e+003 -3.402e+001 -2.843e+003
1120: -1.313e+002 -1.091e+004 1.995e+000 2.051e+002 8.979e+003 7.420e+005 -2.842e+003 -2.388e+005
1121: 3.914e-001 2.967e+001 1.783e-001 1.483e+001 -3.402e+001 -2.842e+003 1.494e+002 1.251e+004
1122: 3.166e+001 2.399e+003 1.486e+001 1.244e+003 -2.843e+003 -2.388e+005 1.251e+004 1.053e+006
1123: # Scales
1124: 12 1.00000e-004 1.00000e-006
1125: 13 1.00000e-004 1.00000e-006
1126: 21 1.00000e-003 1.00000e-005
1127: 23 1.00000e-004 1.00000e-005
1128: # Covariance
1129: 1 5.90661e-001
1130: 2 -7.26732e-003 8.98810e-005
1131: 3 8.80177e-002 -1.12706e-003 5.15824e-001
1132: 4 -1.13082e-003 1.45267e-005 -6.50070e-003 8.23270e-005
1133: 5 9.31265e-003 -1.16106e-004 6.00210e-004 -8.04151e-006 1.75753e+000
1134: 6 -1.15664e-004 1.44850e-006 -7.79995e-006 1.04770e-007 -2.12929e-002 2.59422e-004
1135: 7 1.35103e-003 -1.75392e-005 -6.38237e-004 7.85424e-006 4.02601e-001 -4.86776e-003 1.32682e+000
1136: 8 -1.82421e-005 2.35811e-007 7.75503e-006 -9.58687e-008 -4.86589e-003 5.91641e-005 -1.57767e-002 1.88622e-004
1137: # agemin agemax for lifexpectancy, bage fage (if mle==0 ie no data nor Max likelihood).
1138:
1139:
1140: agemin=70 agemax=100 bage=50 fage=100
1141: Computing prevalence limit: result on file 'plrmypar.txt'
1142: Computing pij: result on file 'pijrmypar.txt'
1143: Computing Health Expectancies: result on file 'ermypar.txt'
1144: Computing Variance-covariance of DFLEs: file 'vrmypar.txt'
1145: Computing Total LEs with variances: file 'trmypar.txt'
1146: Computing Variance-covariance of Prevalence limit: file 'vplrmypar.txt'
1147: End of Imach
1148: </font></pre>
1149: </li>
1.2 lievre 1150: </ul>
1151:
1.6 lievre 1152: <p><font size="3">Once the running is finished, the program
1153: requires a caracter:</font></p>
1.2 lievre 1154:
1.6 lievre 1155: <table border="1">
1.2 lievre 1156: <tr>
1.6 lievre 1157: <td width="100%"><strong>Type e to edit output files, c
1158: to start again, and q for exiting:</strong></td>
1.2 lievre 1159: </tr>
1160: </table>
1161:
1.6 lievre 1162: <p><font size="3">First you should enter <strong>e </strong>to
1163: edit the master file mypar.htm. </font></p>
1164:
1165: <ul>
1166: <li><u>Outputs files</u> <br>
1.3 lievre 1167: <br>
1.6 lievre 1168: - Observed prevalence in each state: <a
1169: href="..\mytry\prmypar.txt">pmypar.txt</a> <br>
1170: - Estimated parameters and the covariance matrix: <a
1171: href="..\mytry\rmypar.txt">rmypar.txt</a> <br>
1172: - Stationary prevalence in each state: <a
1173: href="..\mytry\plrmypar.txt">plrmypar.txt</a> <br>
1174: - Transition probabilities: <a
1175: href="..\mytry\pijrmypar.txt">pijrmypar.txt</a> <br>
1176: - Copy of the parameter file: <a
1177: href="..\mytry\ormypar.txt">ormypar.txt</a> <br>
1178: - Life expectancies by age and initial health status: <a
1179: href="..\mytry\ermypar.txt">ermypar.txt</a> <br>
1.2 lievre 1180: - Variances of life expectancies by age and initial
1.6 lievre 1181: health status: <a href="..\mytry\vrmypar.txt">vrmypar.txt</a>
1.2 lievre 1182: <br>
1.6 lievre 1183: - Health expectancies with their variances: <a
1184: href="..\mytry\trmypar.txt">trmypar.txt</a> <br>
1185: - Standard deviation of stationary prevalence: <a
1186: href="..\mytry\vplrmypar.txt">vplrmypar.txt</a><br>
1187: - Prevalences forecasting: <a href="frmypar.txt">frmypar.txt</a>
1.2 lievre 1188: <br>
1.6 lievre 1189: - Population forecasting (if popforecast=1): <a
1190: href="poprmypar.txt">poprmypar.txt</a> <br>
1191: </li>
1192: <li><u>Graphs</u> <br>
1.2 lievre 1193: <br>
1.6 lievre 1194: -<a href="../mytry/pemypar1.gif">One-step transition
1195: probabilities</a><br>
1196: -<a href="../mytry/pmypar11.gif">Convergence to the
1197: stationary prevalence</a><br>
1198: -<a href="..\mytry\vmypar11.gif">Observed and stationary
1199: prevalence in state (1) with the confident interval</a> <br>
1200: -<a href="..\mytry\vmypar21.gif">Observed and stationary
1201: prevalence in state (2) with the confident interval</a> <br>
1202: -<a href="..\mytry\expmypar11.gif">Health life
1203: expectancies by age and initial health state (1)</a> <br>
1204: -<a href="..\mytry\expmypar21.gif">Health life
1205: expectancies by age and initial health state (2)</a> <br>
1206: -<a href="..\mytry\emypar1.gif">Total life expectancy by
1207: age and health expectancies in states (1) and (2).</a> </li>
1.2 lievre 1208: </ul>
1209:
1.6 lievre 1210: <p>This software have been partly granted by <a
1211: href="http://euroreves.ined.fr">Euro-REVES</a>, a concerted
1.2 lievre 1212: action from the European Union. It will be copyrighted
1213: identically to a GNU software product, i.e. program and software
1214: can be distributed freely for non commercial use. Sources are not
1215: widely distributed today. You can get them by asking us with a
1.6 lievre 1216: simple justification (name, email, institute) <a
1217: href="mailto:brouard@ined.fr">mailto:brouard@ined.fr</a> and <a
1218: href="mailto:lievre@ined.fr">mailto:lievre@ined.fr</a> .</p>
1219:
1.7 brouard 1220: <p>Latest version (0.71a of March 2002) can be accessed at <a
1.8 lievre 1221: href="http://euroreves.ined.fr/imach">http://euroreves.ined.fr/imach</a><br>
1.6 lievre 1222: </p>
1.2 lievre 1223: </body>
1224: </html>
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/icons/small/back.gif){kind=icon}
Return to imach.htm CVS log ![[TXT]](/icons/small/text.gif){kind=icon}
![[DIR]](/icons/small/dir.gif){kind=icon}
Up to [Local Repository] / imach096d / doc