{smcl}
{* MAY 2008}{...}
{hline}
help for {hi:strsrcs}
{hline}
{title:Flexible parametric models for survival-time data with background mortality}
{p 8 16 2}{cmd:strsrcs}{space 2}[{it:varlist}]
[{cmd:if} {it:exp}] [{cmd:in} {it:range}]
[{cmd:,} {it:model_complexity}
{cmdab:bhaz:ard(}{it:varname}{cmd:)}
{cmdab:sc:ale(}{it:hazard|odds}{cmd:)}
{cmdab:strata(}{it:strata_varlist}{cmd:)}
{cmdab:nocons:tant}
{cmdab:snocons:tant}
{cmdab:orthog}
{cmdab:ef:orm}
{cmd:inits(}{it:name}{cmd:)}
{it:maximize_options} ]
{p 4 4 2}where {it:model_complexity} is one of
{p 8 8 2}{c -(} {cmdab:df(}{it:#}{cmd:)}} {cmdab:knots(}{it:knotlist}{cmd:)}} {c )-}
{p 4 4 2}
{cmd:strsrcs} is for use with survival-time data; see help {help st}. You
must {cmd:stset} your data before using {cmd:strsrcs}; see help {help stset}.
{p 4 4 2}
{cmd:strsrcs} shares the features of all estimation commands; see help
{help estcom}.
{p 4 4 2}
The syntax of {help predict} following {cmd:strsrcs} is
{p 8 16 2}{cmd:predict} {it:newvarname} [{cmd:if} {it:exp}]
[{cmd:in} {it:range}] [{cmd:,} {it:statistic} {cmdab:ci} {cmdab:lev:el(}{it:value}{cmd:)}
{p 4 4 2}
where {it:statistic} is
{p 8 28 2}{cmdab:s:urvival} - predicted relative survival
{p 8 28 2}{cmdab:h:azard} - predicted excess hazard rate
{p 8 28 2}{cmdab:hr:atio} - predicted excess hazard rate ratio (NOTE:Only valid when one covariate is modelled)
{p 4 4 2}
These statistics are available both in and out of sample; type
"{cmd:predict} {it:...} {cmd:if e(sample)} {it:...}" if wanted only for the
estimation sample.
Predictions are obtained using the delta method implemented using {help predictnl}.
{title:Description}
{p 4 4 2}
{cmd:strsrcs} fits spline-based distributional models to right censored data taking background mortality into account. varlist is a set of covariates.
{title:Options for {cmd:strsrcs}}
{p 4 8 2}
{cmd:df(}{it:#}{cmd:)} specifies the degrees of freedom for the natural spline function. # must be between 2 and 6. The {cmd:knots()} option is not
applicable and the knots are placed at the following centiles of the distribution of the uncensored event [i.e. where _d==1] log times:
---------------------------
df Centile positions
---------------------------
1 (no knots)
2 50
3 33 67
4 25 50 75
5 20 40 60 80
6 17 33 50 67 83
>6 (not allowed)
---------------------------
{p 4 8 2}
{cmd:knots(}{it:knotlist}{cmd:)} knot placement that defines the internal knot positions for the spline.
The values in knotlist are taken to be centile positions in the distribution of the uncensored event [i.e. where _d==1] log times.
{p 4 8 2}
{cmd:bhazard(}{it:varname}{cmd:)} gives the variable name for the baseline hazard at death/censoring.
This option is compulsary, but a variable containing only zeros can be specified if 'standard' survival models are required as in {help stpm}.
{p 4 8 2}
{cmd:scale(}{it:hazard|odds}{cmd:)} is not optional and specifies the scale of the model. The hazard and odds options fit models on
the scale of the log cumulative hazard or the log cumulative odds of failure, respectively.
{p 4 8 2}
{cmd:strata(}{it:strata_varlist}{cmd:)} stratifies the spline functions according to the variables in strat_varlist.
This allows time-dependant effects for a variety of covariates and and is of particular use in assessing the assumption of proportional excess hazards or proportional odds.
{p 4 8 2}
{cmd:noconstant} specifies that a constant term is not included in the baseline part of the model.
{p 4 8 2}
{cmd:snoconstant} specifies that a constant term is not included for the spline terms.
{p 4 8 2}
{cmd:orthog} creates orthogonalized basis functions. All basis functions higher than the first (linear) function are uncorrelated and have mean 0 and standard deviation 1. The linear function is also uncorrelated with the higher-basis functions.
{p 4 8 2}
{cmd:eform} displays the exponentiated coefficients and corresponding standard errors and confidence intervals.
{p 4 8 2}
{cmd:inits(}{it:name}{cmd:)} specifies a user defined set of initial values, where {it:name} is the name of the matrix where these values are stored.
{p 4 8 2}
{cmd:level(}{it:#}{cmd:)} specifies the confidence level, in percent,
for confidence intervals. The default is {cmd:level(95)} or as set by
{cmd:set level}; see help {help level}.
{p 4 8 2}
{it:maximize_options} control the maximization process; see help
{help maximize}.
{title:Examples}
{p 4 8 2}{cmd:. strsrcs sex, bhazard(rate) df(2) scale(odds) }{p_end}
[assess sex on the odds scale assuming proportional odds]
{p 4 8 2}{cmd:. strsrcs, bhazard(rate) df(2) scale(odds) strata(sex)}{p_end}
[assess sex on the odds scale assuming sex is dependent on time]
{p 4 8 2}{cmd:. strsrcs age, bhaz(rate) df(6) sc(hazard) strata(sex)}{p_end}
[assess age and sex on the hazard scale assuming age has proportional hazards and sex is time dependent]
{p 4 8 2}{cmd:. predict s, survival ci}{p_end}
[output the relative survival estimate (with confidence interval) into new variable {it:s(s_lci & s_uci)}]
{p 4 8 2}{cmd:. predict haz, h ci level(99)}{p_end}
[output the excess hazard (with confidence interval at the 99% level) into new variable {it:haz(haz_lci & haz_uci)}]
{title:Auxiliary ado-files}
rcs strsrcs_mlo strsrcs_mlh strsrcs_pred
{title:Author}
Chris Nelson, University of Leiceser, UK. cn46@le.ac.uk
{title:References}
Nelson, C. P., Lambert, P. C., Squire, I. S. and Jones, D. R. 2007. Flexible parametric models for relative survival, with application
in coronary heart disease. Statistics in Medicine 26: 5486-5498.
Royston, P. 2001. Flexible alternatives to the Cox model, and more. Stata Journal 1: 1-28.
Royston, P. and M. K. B. Parmar. 2002. Flexible proportional-hazards and proportional-odds models for censored survival data, with
application to prognostic modelling and estimation of treatment effects. Statistics in Medicine 21: 2175-2197.
{title:Also see}
{p 4 13 2}
Online: help for {help stpm}, {help strs}, {help estcom}, {help postest};
{help st}, {help stset}