Stata 15 help for stcurve

[ST] stcurve -- Plot survivor, hazard, cumulative hazard, or cumulative incidence function

Syntax

stcurve [, options]

options Description ------------------------------------------------------------------------- Main * survival plot survivor function * hazard plot hazard function * cumhaz plot cumulative hazard function * cif plot cumulative incidence function at(varname=# [varname=# ...]) value of the specified covariates and [at1(varname=# [varname=# ...]) mean of unspecified covariates [at2(varname=# [varname=# ...]) [...]]]

Options alpha1 conditional frailty model fixedonly set all random effects to zero unconditional unconditional frailty model or random-effects model marginal synonym for unconditional range(# #) range of analysis time outfile(filename [, replace]) save values used to plot the curves width(#) override "optimal" width; use with hazard kernel(kernel) kernel function; use with hazard noboundary no boundary correction; use with hazard

Plot connect_options affect rendition of plotted survivor, hazard, or cumulative hazard function

Add plots addplot(plot) add other plots to the generated graph

Y axis, X axis, Titles, Legend, Overall twoway_options any options other than by() documented in [G-3] twoway_options ------------------------------------------------------------------------- * One of survival, hazard, cumhaz, or cif must be specified. survival and hazard are not allowed after estimation with stcrreg. cif is allowed only after estimation with stcrreg.

Menu

Statistics > Survival analysis > Regression models > Plot survivor, hazard, cumulative hazard, or cumulative incidence function

Description

stcurve plots the survivor, hazard, or cumulative hazard function after stcox, streg, stintreg, mestreg, xtstreg. stcurve also plots the cumulative subhazard or cumulative incidence function after stcrreg.

Options

+------+ ----+ Main +-------------------------------------------------------------

survival specifies that the survivor function be plotted. survival is not allowed after estimation with stcrreg.

hazard specifies that the hazard function be plotted. hazard is not allowed after estimation with stcrreg.

cumhaz specifies that the cumulative hazard function be plotted when used after stcox, streg, stintreg, mestreg, or xtstreg and specifies that the cumulative subhazard function be plotted when used after stcrreg.

cif specifies that the cumulative incidence function be plotted. This option is available only after estimation with stcrreg.

at(varname=# ...) specifies that the covariates specified by varname be set to #. By default, stcurve 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.

at1(varname=# ...), at2(varname=# ...), ..., at10(varname=# ...) specify that multiple curves (up to 10) be plotted on the same graph. at1(), at2(), ..., at10() work like the at() option. They specify that the function be evaluated at the value of the covariates specified and at the mean of all unlisted covariates. at1() specifies the values of the covariates for the first curve, at2() specifies the values of the covariates for the second curve, and so on.

+---------+ ----+ Options +----------------------------------------------------------

alpha1, when used after fitting a frailty model, plots curves that are conditional on a frailty value of one. This is the default for shared-frailty models.

fixedonly specifies that all random effects be set to zero, which is equivalent to using only the fixed portion of the model, when plotting results for random-effects models. This option is allowed only after xtstreg or mestreg; it is the default after xtstreg.

unconditional and marginal, when used after fitting a frailty model or a random-effects model, plot curves that are unconditional on the frailty or on the random effects. That is, the curve is "averaged" over the frailty distribution or over the random-effects distributions. This is the default for unshared-frailty models and for random-effects models. This option is not allowed after stintreg or xtstreg.

range(# #) specifies the range of the time axis to be plotted. If this option is not specified, stcurve plots the desired curve on an interval expanding from the earliest to the latest time in the data.

outfile(filename [, replace]) saves in filename.dta the values used to plot the curve(s).

width(#) is for use with hazard (and applies only after stcox) and is used to specify the bandwidth to be used in the kernel smooth used to plot the estimated hazard function. If left unspecified, a default bandwidth is used, as described in [R] kdensity.

kernel(kernel) is for use with hazard and is for use only after stcox because, for Cox regression, an estimate of the hazard function is obtained by smoothing the estimated hazard contributions. kernel() specifies the kernel function for use in calculating the weighted kernel-density estimate required to produce a smoothed hazard-function estimator. The default is kernel(epanechnikov), yet kernel may be any of the kernels supported by kdensity; [R] kdensity.

noboundary is for use with hazard and applies only to the plotting of smoothed hazard functions after stcox. It specifies that no boundary-bias adjustments are to be made when calculating the smoothed hazard-function estimator. By default, the smoothed hazards are adjusted near the boundaries; see [ST] sts graph. If the epan2, biweight, or rectangular kernel is used, the bias correction near the boundary is performed using boundary kernels. For other kernels, the plotted range of the smoothed hazard function is restricted to be inside of one bandwidth from each endpoint. For these other kernels, specifying noboundary merely removes this range restriction.

+------+ ----+ Plot +-------------------------------------------------------------

connect_options affect the rendition of the plotted survivor, hazard, or cumulative hazard function; see [G-3] connect_options.

+-----------+ ----+ Add plots +--------------------------------------------------------

addplot(plot) provides a way to add other plots to the generated graph; see [G-3] addplot_option.

+-----------------------------------------+ ----+ Y axis, X axis, Titles, Legend, Overall +--------------------------

twoway_options are any of the options documented in [G-3] twoway_options, excluding by(). These include options for titling the graph (see [G-3] title_options) and for saving the graph to disk (see [G-3] saving_option).

Examples after stcox

Setup . webuse drugtr

Fit Cox model . stcox age drug

Plot the estimated survivor function . stcurve, survival

Plot the estimated survivor function for the placebo group and for the treatment group . stcurve, survival at1(drug=0) at2(drug=1)

Plot the estimated hazard function for the placebo group and for the treatment group on a log scale . stcurve, hazard at1(drug=0) at2(drug=1) kernel(gauss) yscale(log)

Examples after streg

--------------------------------------------------------------------------- Setup . webuse cancer

Map values for drug into 0 for placebo and 1 for nonplacebo . replace drug = drug == 2 | drug == 3

Fit a loglogistic survival model . streg age drug, d(llog)

Plot the survivor function . stcurve, survival ylabels(0 .5 1)

Plot the hazard function . stcurve, hazard

Plot the survivor function for the placebo group and for the treatment group . stcurve, survival at1(drug=0) at2(drug=1) ylabels(0 .5 1)

--------------------------------------------------------------------------- Setup . webuse catheter, clear

Declare data to be survival-time data . stset time infect

Fit a Weibull/inverse-Gaussian shared-frailty model . streg age female, d(weibull) frailty(invgauss) shared(patient)

Plot the individual hazard function for females at mean age . stcurve, hazard at(female=1) alpha1

Plot the population hazard function for females at mean age . stcurve, hazard at(female=1) unconditional

---------------------------------------------------------------------------

Example after stcrreg

Setup . webuse hypoxia

Declare data to be survival-time data . stset dftime, failure(failtype==1)

Fit competing-risks model . stcrreg ifp tumsize pelnode, compete(failtype==2)

Compare cumulative incidence functions for two groups with remaining covariates held at their mean values . stcurve, cif at1(pelnode=0) at2(pelnode=1)

Example after mestreg

Setup . webuse catheter

Random-effects Weibull survival model . mestreg age female || patient:, distribution(weibull)

Compare marginal survivor functions for males and females at age 53 . stcurve, surv at1(female=0 age=53) at2(female=1 age=53) Example after xtstreg

Setup . webuse catheter . xtset patient

Random-effects Weibull survival model . xtstreg age female, distribution(weibull)

Compare survivor functions for males and females at mean age conditional on zero random effects . stcurve, surv at1(female=0) at2(female=1)

Example after stintreg

Setup . webuse cosmesis

Fit a Weibull survival model for interval-censored data . stintreg i.treat, distribution(weibull) interval(ltime rtime)

Compare survivor functions for two treatments: radiotherapy only and radiation plus chemotherapy . stcurve, surv at1(treat=0) at2(treat=1)


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index