beanieclo@HOTMAIL.COM wrote: >I found this very helpful page on how to compute robust test statistics >for the coefficients using the consistent variance matrix (obtained >with the ACOV function in PROC REG): > >http://www.ats.ucla.edu/stat/sas/webbooks/reg/chapter4/sasreg4.htm

The whole site is really useful. I recommend that people start at:

http://www.ats.ucla.edu/stat/sas/

HOWEVER I do have one complaint about the webpage that Claudia has pointed out. The authors go to the trouble of figuring out that the data come from a sample with clustering, and so they propose PROC GENMOD to address the clustering effect. Even though this is now obviously some sort of survey sample! This is clearly the sort of problem for which PROC SURVEYREG was designed.

I gnashed my teeth too much while writing this. I must now go and get some soothing chocolate ice cream. Preferably from Ben and/or Jerry. :-) :-)

David -- David L. Cassell mathematical statistician Design Pathways 3115 NW Norwood Pl. Corvallis OR 97330

_________________________________________________________________ On the road to retirement? Check out MSN Life Events for advice on how to get there! http://lifeevents.msn.com/category.aspx?cid=Retirement