Date:         Sun, 17 May 2009 13:53:48 -0700
Reply-To:     Ryan <Ryan.Andrew.Black@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Ryan <Ryan.Andrew.Black@GMAIL.COM>
Organization: http://groups.google.com
Subject:      Random effects hurdle model  - conceptual question
Comments: To: sas-l@uga.edu
Content-Type: text/plain; charset=ISO-8859-1

Hi,

I have a conceptual question about a random effects hurdle model.

Dataset:

/************************/

Person   Group     Y
1              1         0
1              2         2
1              3         0
2              1         0
2              2         0
2              3        15
.
.
.
/***********************/

Would one need to account for person random effects on the count (e.g.
Poisson) equation, if most people (>95%) only have a non-zero value
for one level of "Group"? It seems to me that since the count (e.g.
Poisson) equation is based solely on non-zero observations, there
would be no need to account for person random effects.

Am I correct on this?

Thanks!