Prev: please help
Next: Cluster Analysis with both characteristic and numeric variables
From: Pakcha Sharma on 7 Sep 2009 23:13
Hi all,

I am trying to model a non-negative continuous variable, which is highly
right skewed and has a lot of zero values (more than 30%), and analyse its
association with other independent variables. I have tried PROC GENMOD
with gamma distribution but the analyse drops all zero values from the
analysis. I read it somewhere that I can fit a zero inflated gamma
distribution.

I have been trying to find a reference for that and a SAS procedure that
can do it but have been unable to find. Could some one please email me the
SAS procedure and a reference or point me to the right direction.

Thanks.
 | 
Pages: 1
Prev: please help
Next: Cluster Analysis with both characteristic and numeric variables