3.8. Predicted Values, Residuals, and Influence Statistics
Next, we turn to diagnostic statistics that are computed for each individual observation. GENMOD and LOGISTIC produce a large number of case-wise statistics that can either be printed in the output display or written to a separate file. Here are the ones produced by both procedures:
Linear predictor— Predicted log-odds for each case. In matrix notation, this is xβ, so it’s commonly referred to as XBETA.
Standard error of linear predictor— Used in generating confidence intervals.
Predicted values— Predicted frequency of the event, based on the estimated model and values of the explanatory variables. For individual-level data, this is a predicted probability.
Confidence intervals for predicted ...
Get Logistic Regression Using SAS®: Theory and Application now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
