I have Q3 sales for consecutive years 2002-2007 that I'm using to predict buying (yes/no) in Q3 in 2008. My data are arranged in counting process format where each customer has 6 rows of data, one for each year. However, there is no variability within the strata at the customer level: if someone bought in 2008, then all 6 records will be flagged as buy=1. I don't think this will work and I can't figure out how to get around this in order account for the serial correlation in the predictors. Can someone point me to the correct analytical technique? Is there an R module that will perform this analysis? [ALERT] -- Access Manager: This email is intended only for the person or entity to which it is addressed and may contain information that is privileged, confidential or otherwise protected from disclosure. Dissemination, distribution or copying of this e-mail or the information herein by anyone other than the intended recipient, or an employee or agent responsible for delivering the message to the intended recipient, is prohibited. If you have received this e-mail in error, please immediately notify us by calling our North American Help Desk at (972)506-3939. Targetbase Messaging Services provided by DMSP [[alternative HTML version deleted]]