Chapter Contents |
Previous |
Next |

The LIFEREG Procedure |

For each model, PROC LIFEREG displays

- the name of the Data Set
- the name of the Dependent Variable
- the name of the Censoring Variable
- the Censoring Value(s) that indicate a censored observation
- the number of Noncensored and Censored Values
- the final estimate of the maximized log likelihood
- the iteration history and the Last Evaluation of the Gradient and Hessian if the ITPRINT option is specified (not shown)

For each explanatory variable in the model, the LIFEREG procedure displays

- the name of the Variable
- the degrees of freedom (DF) associated with the variable in the model
- the Estimate of the parameter
- the standard error (Std Err) estimate from the observed information matrix
- an approximate chi-square statistic for testing that the parameter is zero (the class variables also have an overall chi-square test statistic computed that precedes the individual level parameters)
- the probability of a larger chi-square value (Pr>Chi)
- the Label of the variable or, if the variable is a class level, the Value of the class variable

If there are constrained parameters in the model, such as the scale or intercept, then PROC LIFEREG displays a Lagrange multiplier test for the constraint.

Chapter Contents |
Previous |
Next |
Top |

Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.