Help using Base SAS procedures

Proc GENMOD output

Reply
Occasional Contributor
Posts: 7

Proc GENMOD output

i am exploring SAS Enterprise Miner right now. i am working with proc Genmod. 

 

 

Analysis Of Maximum Likelihood Parameter Estimates
 
                                       Standard     Wald 95% Confidence
Parameter            DF    Estimate       Error           Limits

Intercept             1     -2.5659      0.0666     -2.6964     -2.4354
Brand        PBB      1      1.0135      0.0804      0.8559      1.1711
Brand        PFl      1     -0.2243      0.0991     -0.4185     -0.0301
Brand        PGen     1      0.2179      0.0901      0.0413      0.3945
Brand        PHse     1      0.9059      0.0814      0.7465      1.0654
Brand        PPk      1      2.3614      0.0750      2.2145      2.5083
Brand        PSS      0      0.0000      0.0000      0.0000      0.0000
Scale                 0      1.0000      0.0000      1.0000      1.0000

NOTE: The scale parameter was held fixed.

to my understanding, i am reading the output like, for example for Brand PPk estimate is 2.3614. that means this brand is 2.3614 likely to be the choice than the other 5. and for brand PSS its 0.0000 that means it was never choosed.

 

Am i right in onterpreting the results??? 

Occasional Contributor
Posts: 7

Re: Proc GENMOD output

Posterior Correlation Matrix
 
                               Brand       Brand       Brand       Brand       Brand
Parameter      Intercept         PBB         PFl        PGen        PHse         PPk

Intercept          1.000      -0.829      -0.688      -0.738      -0.820      -0.888
BrandPBB          -0.829       1.000       0.566       0.617       0.676       0.742
BrandPFl          -0.688       0.566       1.000       0.510       0.561       0.609
BrandPGen         -0.738       0.617       0.510       1.000       0.594       0.653
BrandPHse         -0.820       0.676       0.561       0.594       1.000       0.727
BrandPPk          -0.888       0.742       0.609       0.653       0.727       1.000

for this posterior corelation matrix, output will be interprted as for example, Brand PBB and Brand PPk are strongly corelated as 0.742 is close to 1. ????

Ask a Question
Discussion stats
  • 1 reply
  • 145 views
  • 0 likes
  • 1 in conversation