Chapter 73: The TPHREG Procedure (Experimental)


Overview

The TPHREG procedure, experimental in this release, includes most of the functionality of the PHREG procedure (see Chapter 54, 'The PHREG Procedure') with the additional benefits of the CLASS statement. The CLASS statement enables you to specify categorical variables (also known as factors or CLASS variables) to be used in the analysis. Model effects, including covariates, main effects (CLASS variables ), crossed effects (interactions), and nested effects, can be specified in the same way as in the GLM procedure. The CLASS statement supports the less-than -full-rank parameterization as in the GLM procedure as well as various full-rank parameterization methods such as reference coding, effect coding, and orthogonal polynomial coding. For some of the full-rank coding schemes, you can designate a specific value (category or level) of the CLASS variable as the reference level. The CLASS statement also enables you to specify the ordering of the categories of CLASS variables, to reverse the ordering of the categories, and to treat categories with missing values as valid categories.

With the TPHREG procedure, you can control how to move model effects in and out of a model with various model-building strategies such as forward selection, backward elimination , or stepwise selection. When there are no interaction terms, a main effect can enter or leave a model in a single step based on the p -value of the score or Wald statistic, respectively. When there are crossed or nested effects, the selection process also depends on whether you want to preserve model hierarchy. The HIERARCHY= option in the MODEL statement enables you to specify whether model hierarchy is to be preserved, how model hierarchy is applied, and whether a single effect or multiple effects can be moved in a single step.

The TPHREG procedure also enables you to specify CONTRAST statements for testing customized hypotheses concerning the regression parameters. Each CONTRAST statement also provides estimation of individual rows of contrasts, which is particularly useful in comparing the hazards between the categories of a CLASS variable.




SAS.STAT 9.1 Users Guide (Vol. 6)
SAS.STAT 9.1 Users Guide (Vol. 6)
ISBN: N/A
EAN: N/A
Year: 2004
Pages: 127

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net