15.7. Using a Modified Wald Statistic to Assess Model Effects
Section 15.3 discusses the possibility of using an adjusted Wald statistic to evaluate model effects in the GEE approach. Shah, Holt, and Folsom (1977) describe a modification of the Wald statistic based on a Hotelling T2 type of transformation of QC:
The quantity d is equal to the number of clusters minus 1 and c is equal to the number of rows of the contrast. Thus, for tests concerning effects of explanatory factors, it is equal to the corresponding number of df. This test is more conservative than the Wald test; LaVange, Koch, and Schwartz (2000) suggest that you can use the Wald ...
Get Categorical Data Analysis Using The SAS® System, 2nd Edition 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.
