91
Complex Samples Cox Regression
iteration(the i nitial estimates), where nis the value of theincrement. If the iterationhistory is
requested, then the lastiteration is always displayed regardless of n.
Tiebreaking method for parameter estimation. Whenthere are tied observed failure times, one of
thesemethods is used to break the ties. TheEfron method is more computationally expensive.
SurvivalFunctions. Thesecontrols specify criteria for computations inv olving the survival function.
Methodfor estimating baseline survival functions. The Breslow(orNelson-Aalanorempirical)
method estimates the baseline cumulativehazard by a nondecreasing step function with
steps at the observed failure times, then computes the baselinesurvival by the relation
survival=exp(−cumulativehazard). The Efron method is more computationally expensive
and reduces to the Breslow method when there are no ties. The product limit method
estimates the baselinesurvival by a non-increasing right continuous function; when there are
no predictors in the model,this method reduces to Kaplan-Meier estimation.
Confidenceintervals of su rvivalfunctions. The confidence interval can be calculated in three
ways: in original units, via a log transformation, or a log-minus-log transformation. Only the
log-minus-logtransformation guarantees that the bounds of the confidence interval will lie
between 0 and 1, butthe log transformation generally seems to perform “best.”
UserMissing Values. Allvariables must have valid values for a case to be included in the analysis.
These controls allow you to decide whether user-missingvalues are treated as valid among
categorical models (including factors,event, strata, and subpopulation variables) and sampling
design variables.
Confidence interval(%). This is the confidence interval level used fo r coefficientestimates,
exponentiated coefficient estimates, survival function estimates, and cumulativehazard function
estimates. Specify a value greater than or equal to 0, and less than 100.
CSCOXREG Command Additional Features
The command language also allowsyou to:
Perform custom hypothesis tests (using the CUSTOM subcommand and /PRINT LMATRIX).
Toler ance sp ecification (using /CRITERIA SINGULAR).
General estimable functiontable (using /PRINT GEF).
Multiple predictor patterns (using multiplePATTERN subcommands).
Maximum number of saved variables when a rootname is specified(u sing the SAVE
subcommand). The dialog honors the CSCOXREG default of 25 variables.
See the Command Syntax Referencefor complete syntax information.