| Poisson Regression |
The Type III (Wald) Tests table in this example is a further breakdown of the variation due to MODEL. The DF for Model are broken down into terms corresponding to the main effects for YEAR, TYPE, and PERIOD, and the interaction effects for TYPE*YEAR, YEAR*PERIOD, and TYPE*PERIOD. The composite explanatory power of the set of parameters associated with each effect is measured by the Chi-Square statistic. The p-value corresponding to each Chi-Square statistic is the probability of observing a statistic of equal or greater value, given that the corresponding parameters are all 0.
Copyright © 2007 by SAS Institute Inc., Cary, NC, USA. All rights reserved.