BookmarkSubscribeRSS Feed
Calcite | Level 5

I was trying to fit neg binomial and poisson distribution parameters to a single data set (no predictors just a set of values) using Maximun Likelihood Estimation and then conduct a likelihood ratio test to see if the neg binomial fits the data better than the poisson.  The problem is that I only have SAS Base 9.1 and SAS Stat, I don’t have SAS IML.  The only option that I have was working with SAS Genmode procedure, but this is only for regression models.  However, I found two things running the procedure with my data set as response variable and with no predictors.  First, the dispersion parameter displayed in the “Analysis Of Maximum Likelihood Parameter Estimates” table for the negative binomial, gives the same number, for the distribution parameter r, I get using other methods (in R) that are meant to work with a single data set (no regression analysis).  Second, I calculate the test statistic of the likelihood ratio test, using the loglikelihood values displayed in the genmode output and I get the same value for the test statistic too.  My question is why this happens and can I make my distributional analysis using the genmode procedure in this way (with no predictors)? 

SAS Employee

Hi Agnes,

Sorry, but I don't think we can help you here in this forum, it is about SAS/OR and it sounds like you don't have that available either. Your best bet is probably to repost your question in the SAS Statistical Procedures forum.




Time is running out to save with the early bird rate. Register by Friday, March 1 for just $695 - $100 off the standard rate.


Check out the agenda and get ready for a jam-packed event featuring workshops, super demos, breakout sessions, roundtables, inspiring keynotes and incredible networking events. 


Register now!

Multiple Linear Regression in SAS

Learn how to run multiple linear regression models with and without interactions, presented by SAS user Alex Chaplin.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 1 reply
  • 2 in conversation