The SURVEYPHREG Procedure
- Overview
- Getting Started
-
Syntax
PROC SURVEYPHREG StatementBY StatementCLASS StatementCLUSTER StatementDOMAIN StatementESTIMATE StatementFREQ StatementLSMEANS StatementLSMESTIMATE StatementMODEL StatementNLOPTIONS StatementOUTPUT StatementProgramming StatementsREPWEIGHTS StatementSLICE StatementSTORE StatementSTRATA StatementTEST StatementWEIGHT Statement -
DetailsNotation and EstimationFailure Time DistributionTime and CLASS Variables UsagePartial Likelihood Function for the Cox ModelSpecifying the Sample DesignMissing ValuesVariance EstimationDomain AnalysisHypothesis Tests, Confidence Intervals, and ResidualsOutput Data SetsDisplayed OutputODS Table NamesODS Graphics
-
Examples
- References
Confidence Intervals
By default, the SURVEYPHREG procedure computes t confidence limits for the estimated regression coefficients. Alternatively, you can specify DF=NONE
in the MODEL statement to request standard normal confidence intervals. The t confidence interval for a linear combination 
of the regression coefficients is computed as
![\[ \left( \mb{l}’ \hat{\bbeta } \pm t_{\mi{df},\alpha /2} \sqrt {\mb{l}’\hat{\mb{V}}(\hat{\bbeta })\mb{l}} \right) \]](images/statug_surveyphreg0141.png)
where 
is the 
percentile point of the t distribution with df degrees of freedom. See the section Degrees of Freedom for more information about df. If you use the DF=NONE
option in the MODEL statement, then the procedure uses the percentile point of the standard normal distribution.