.-
help for ^stcoxgof^
.-
Perform goodness-of-fit test and plot after a proportional hazards model
------------------------------------------------------------------------
^stcoxgof^ [^,^ ^gr^oup^(^#^)^ ^gen(^string^)^ ^poi^dis ^plot(^#^)^
^sa^ving^(^filename^)^ graph_options ]
^stcoxgof^ is for use after estimation of a proportional hazards model.
If used after ^stcox^ this command requires that you previously specified
^stcox^'s ^mgale()^ option; see help @stcox@.
It can be used after a parametric proportional hazards model as well. Then
^streg^ with ^exponential^, ^weibull^ or ^gompertz^ distribution has to be
previously specified; see help @streg@.
Description
-----------
When used without options ^stcoxgof^ shows a two by ten table presenting
the observed and expected numbers of events in each decile of risk.
According to May and Hosmer z-score and two-tailed p-value from standard
normal distrubution are also tabulated. If ^plot^ option is given Arjas
plots by quantiles of risk are displayed.
Options
-------
^group(^#^)^ specifies the number of quantiles of risk to be used to group
observed ed expected numbers of events. Default is 10. Values allowed
are from 4 to 10.
^gen(^string^)^ supplies the name of a new variable to hold the quantiles of
risk indicators. Including this variable as design variables you can
estimate likelihood ratio test for the fitted proportional hazard model.
^poidis^ estimates the probability of observed counts within each decile of
risk according to Poisson distibution with mean equal to the estimated
expected number of counts.
^plot(^#^)^ specifies the number of quantiles of risk used to group the
data for Arjas plots. Values allowed are from 2 to 10. If more than 6
quantiles of risk are used, two combined plots are displayed. To
implement this option ^forgraph^ (STB-49: gr36) is called. See help
@net@ to install user-written additions from the net.
^separate^ requests that for each quantile of risk a separate graph should
be shown.
^saving(^filename^,^[^replace^]^)^ saves graph(s) in file(s) that can be
reviewed by ^graph using^ and printed.
Examples
--------
. ^stcoxgof^
. ^stcoxgof, gr(5)^
. ^stcoxgof, plot(4)^
. ^stcoxgof, gen(decile)^
. ^stcoxgof, poidis^
. ^stcoxgof, plot(8) sa(Arjas,replace)^
Also see
--------
Manual: ^[R] stcox^
On-line: help for @stcox@;
Reference
---------
Hosmer DW,jr and Lemeshow S: Applied Survival Analysis (1999) p. 225 - 230.
John Wiley & Sons - New York.