Model Fitting: Generalized Linear Models: Output Variables Tab

Output Variables Tab

You can use the Output Variables tab to add analysis variables to the data table. (See Figure 24.24.) If you request a plot that uses one of the output variables, then that variable is automatically created even if you did not explicitly select the variable on the Output Variables tab.

For a multinomial response, residuals and influence diagnostics are not available.

The following list describes each output variable and indicates how the output variable is named. $\text{[math]}$ represents the name of the response variable. If you use events/trials syntax, then $\text{[math]}$ represents the name of the events variable.

Proportions for events/trials: adds a variable named Proportion_ $\text{[math]}$ , where $\text{[math]}$ is the name of the events variable and $\text{[math]}$ is the name of the trials variable. The value of the variable is the ratio $\text{[math]}$ . This variable is added only when you use events/trials syntax.
Predicted values: adds predicted values. The variable is named GenP_ $\text{[math]}$ .
Confidence limits for predicted values: adds 95% confidence limits for the predicted values. The variables are named GenLclm_ $\text{[math]}$ and GenUclm_ $\text{[math]}$ .
Linear predictor: adds the linear predictor values. The variable is named GenXBeta_ $\text{[math]}$ .
Raw residuals: adds residuals, which are calculated as observed values minus predicted values. The variable is named GenR_ $\text{[math]}$ .
Pearson chi-square residuals: adds the Pearson chi-square residuals. The variable is named GenChiSqR_ $\text{[math]}$ .
Deviance residuals: adds the deviance residuals. The variable is named GenDevR_ $\text{[math]}$ .
Likelihood residuals: adds the likelihood residuals. The variable is named GenLikR_ $\text{[math]}$ .
Cook’s D: adds Cook’s $\text{[math]}$ influence statistic. The variable is named GenCooksD_ $\text{[math]}$ .
Leverage (H): adds the leverage statistic. The variable is named GenH_ $\text{[math]}$ .
DFBETAS (influence on coefficients): adds $\text{[math]}$ variables, where $\text{[math]}$ is the number of parameters in the model. A classification variable with $\text{[math]}$ levels counts as $\text{[math]}$ parameters. The variables are scaled measures of the change in each parameter estimate and are calculated by deleting the $\text{[math]}$ th observation. Large values of DFBETAS indicate observations that are influential in estimating a given parameter. Belsley, Kuh, and Welsch (1980) recommend $\text{[math]}$ as a size-adjusted cutoff. The variables are named DFBeta $\text{[math]}$ , for $\text{[math]}$ .

Figure 24.24 The Output Variables Tab