The SURVEYPHREG Procedure
- Overview
- Getting Started
-
Syntax
PROC SURVEYPHREG Statement BY Statement CLASS Statement CLUSTER Statement DOMAIN Statement ESTIMATE Statement FREQ Statement LSMEANS Statement LSMESTIMATE Statement MODEL Statement NLOPTIONS Statement OUTPUT Statement Programming Statements REPWEIGHTS Statement SLICE Statement STORE Statement STRATA Statement TEST Statement WEIGHT Statement -
Details
Notation and Estimation Failure Time Distribution Time and CLASS Variables Usage Partial Likelihood Function for the Cox Model Specifying the Sample Design Missing Values Variance Estimation Domain Analysis Hypothesis Tests, Confidence Intervals, and Residuals Output Data Sets Displayed Output ODS Table Names ODS Graphics
-
Examples
- References
Variance Adjustment Factors
PROC SURVEYPHREG provides options for adjustment of the default variance estimators. VADJUST=NONE and VADJUST=DF are available for the Taylor series linearization variance estimator. VADJUST=AVGREPSS is available for the jackknife replication variance estimators.
For models with large number of parameters, it is reasonable to adjust the Taylor series linearized variance estimator by the number of estimable parameters in the analysis model. Fuller et al. (1989, pp. 77–81) use an adjustment factor 
to estimate the linearized variance for regression coefficients, where 
is the total number of observation units and 
is the number of estimable parameters in the analysis model. By default, PROC SURVEYPHREG uses this adjustment in the computation of the matrix 
for the Taylor series linearization variance estimation. If you do not want to use this adjustment, then specify VADJUST=NONE.
Variance adjustment factors can be useful for replication variance estimations, especially if some replicate samples are not usable. A replicate sample might not provide useful parameter estimates (replicate estimates) for reasons such as nonconvergence of the optimization or inestimability of some parameters in that subsample. For example, consider the jackknife variance estimator with 
replicates. Suppose that only 
replicates are used to obtain replicate estimates and 
replicates cannot be used due to, say, nonconvergence of the optimization. Without loss of generality, assume that the first replicates are used. By default SURVEYPHREG uses
 |
as the jackknife variance estimator. An alternative estimator is
 |
 |
 |
|||
 |
|
 |
which uses the average replicate sum of squares for the unusable replicate samples. If you specify the VADJUST=AVGREPSS option, PROC SURVEYPHREG uses the second variance estimator for the jackknife replication method. Note that you can specify the FAY method-option for the BRR method to avoid nonconvergence of the optimization or inestimability of some parameters in subsamples.