Title
igencox postestimation -- Postestimation tools for igencox
The following postestimation commands are available after igencox:
Command Description ------------------------------------------------------------------------- INCLUDE help post_estat INCLUDE help post_estimates INCLUDE help post_lincom INCLUDE help post_lrtest INCLUDE help post_nlcom predict predicted values INCLUDE help post_predictnl INCLUDE help post_test INCLUDE help post_testnl -------------------------------------------------------------------------
predict [type] newvar [if] [in] [, options]
statistic Description ------------------------------------------------------------------------- Main xb linear prediction xb; the default stdp standard error of the linear prediction; SE(xb) basesurv baseline survivor function survival covariate-adjusted survivor function basechazard baseline cumulative hazard function cumhaz covariate-adjusted cumulative hazard function * at(varname=# [varname=# ...]) set value of the specified covariates + se(newvarname) save standard error in newvarname ------------------------------------------------------------------------- * at() is allowed only with survival or cumhaz. + se() is allowed only with basesurv or survival.
+------+ ----+ Main +-------------------------------------------------------------
xb, the default, calculates the linear prediction from the fitted model. That is, you fit the model by estimating a set of parameters b0, b1, b2, ..., bk, and the linear prediction is xb.
The x used in the calculation is obtained from the data currently in memory and need not correspond to the data on the independent variables used in estimating b.
stdp calculates the standard error of the prediction, that is, the standard error of xb.
basesurv calculates the baseline survivor function. It requires that the baseq() option is specified with igencox.
survival calculates the covariate-adjusted survivor function. If at() is not specified, the survivor function is evaluated at the mean values of the covariates. survival requires that the baseq() option is specified with igencox.
basechazard calculates the cumulative baseline hazard. It requires that the baseq() option is specified with igencox.
cumhaz calculates the covariate-adjusted cumulative hazard. If at() is not specified, the cumulative hazard is evaluated at the mean values of the covariates. cumhaz requires that the baseq() option is specified with igencox.
at(varname=# ...) requests that the covariates specified by varname be set to # when computing covariate-adjusted survivor or cumulative hazard functions. By default, predict evaluates the function by setting each covariate to its mean value. This option causes the function to be evaluated at the value of the covariates listed in at() and at the mean of all unlisted covariates. The at() option requires survival or cumhaz.
se(newvarname) requests that the standard error of the survivor function be saved in newvarname. It requires that baseq() and savesigma() are specified with igencox. The se() option requires basesurv or survival.
Compute baseline survivor function
Setup . use va, clear
Fit Cox proportional hazards model and save jump sizes . igencox status type1 type2 type3, baseq(q)
Obtain predicted values of the baseline survivor function . predict s, basesurv
Compute covariate-adjusted survivor function and its standard error
Setup . use va, clear . igencox status type1 type2 type3, baseq(bq) savesigma(tmp, replace)
Predict survivor function and its standard errors at specified values of the covariates . predict surv, survival se(sesurv) at(status=.4 type1=0 type2=0 type3=0)
Calculate the 95% pointwise confidence intervals of the survivor function . gen tmp = 1.96*sesurv / (surv*log(surv)) . gen lb = surv^exp(-tmp) . gen ub = surv^exp(tmp)
Plot the results . gsort _t surv . twoway line surv lb ub _t, connect(J J J)