04-22-2015 02:16 PM
My question is regarding the modified possion regression with robust variance estimation. The performance of this method is verified for independent data (Guangyong Zou 2003) and clustered data using GEE to account for clustering (Lisa Yelland etal 2011). Has the performance also been verified for mixed-effects model or repeated measurements taken on the same individuals over time? If so, detailed literature information will be appreciated. And, to build this model in SAS using the glimmix procedure, is it just including emprical in the proc glimmix statement?