Regression Results - Advanced Tab

Select the Advanced tab of the Regression Results dialog box to access the following options. Use these options to view the parameter estimates, covariances, and correlations, the results of the stratified analysis, and the means and standard deviations.
Option Description
Summary: Parameter estimates Displays a spreadsheet with parameter estimates, their standard errors, the ratio of the parameter estimates divided by their standard errors (under the heading t-value), and the Wald statistic (which for large samples has a Chi-square distribution under the null hypothesis that the coefficient is equal to 0).
Note: The standard errors (and t-values) are computed from the second order partial derivatives of the log-likelihood function, resulting in asymptotic standard errors (and t-values) for the parameters.
Parameter covariances & correlations Displays two spreadsheets, one with the variance/covariance matrix of the parameter estimates and one with the correlation matrix.
Stratified analysis Displays a spreadsheet with the results of the stratified analysis; specifically, the spreadsheet shows the sum of the log-likelihoods of the model, fitted within each group, and the log-likelihood of the model fitted to all observations combined. In order to examine the parameter estimates for a particular group, specify the appropriate case selection conditions, that is, select only cases belonging to the group of interest, and then fit the respective model in the usual manner.

Stratified analysis is available only if a stratified analysis (analysis by groups) is requested on the Regression Models for Censored Data dialog box - Quick tab or the Proportional Hazard Model with Time-Dependent Covariates dialog box - Quick tab (that is, a grouping variable has been selected) and if the Equal coefficients, different baseline h0 (in stratified analysis) check box on the Regression Models for Censored Data dialog box - Options tab is not selected.

Means & standard deviations Displays a spreadsheet with the means and standard deviations for the survival times and the independent variables (covariates).