Date:         Wed, 28 Nov 2007 09:14:15 -0800
Reply-To:     jn mao <jn_mao@YAHOO.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         jn mao <jn_mao@YAHOO.COM>
Subject:      Re: Poisson model and NB model
Comments: To: Peter Flom <peterflomconsulting@mindspring.com>
In-Reply-To:  <1004129.1196269449686.JavaMail.root@mswamui-backed.atl.sa.earthlink.net>
Content-Type: text/plain; charset=iso-8859-1

The adjusted code for overdispersed poisson model is:
  proc genmod;
  class var1;
  model dependent=var1 var2 var3 /link=log dist=poisson offset=followln scale=deviance;

  The NB model code is;
  proc genmod;
  class var1;
  model dependent=var1 var2 var3 /link=log dist=NB offset=followln;

  the var is signficant in overdispersed poisson model, but not significant in NB model.

  Jane

Peter Flom <peterflomconsulting@mindspring.com> wrote:
  jn mao wrote

>Hello SAS-Ls,
>
> Could someone give me some advice about how to choose adjusted Poisson model and NB model?
> I'm analyzing an over-dispersed data, the NB model shown one variable is not significant, but when I used scale deviance adjust poisson model, that variable becomes significant. I'm not sure which model I should use? why the adjusted poisson model does not consist wit£è¡¡
> NB model?
>

What adjusted Poisson model is this?
Can you show code, or describe it some way?

thanks

Peter



---------------------------------
Never miss a thing.   Make Yahoo your homepage.