turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Z-test in NB regression model

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

06-21-2011 01:06 PM

I modeled a NB regression as below

proc genmod data=Crash; class ID year;

model accident= major minor/dist=nb link=log

type3 wald; repeated subject=ID;

run; quit;

the result shows Z test even that I said wald and no overdispersion is shown. If I remove "repeated subject=ID" then it shows wald chi square test and the overdispersion.

Is there any way I can get chi square test result and over dispersion? ( I don't even need it to be wald, even If I don't type wald it still shows me wald chi square)

Thanks

proc genmod data=Crash; class ID year;

model accident= major minor/dist=nb link=log

type3 wald; repeated subject=ID;

run; quit;

the result shows Z test even that I said wald and no overdispersion is shown. If I remove "repeated subject=ID" then it shows wald chi square test and the overdispersion.

Is there any way I can get chi square test result and over dispersion? ( I don't even need it to be wald, even If I don't type wald it still shows me wald chi square)

Thanks

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to Najmeh

06-24-2011 01:11 PM

The WALD option in the MODEL statement causes the tests produced by the TYPE3 option to be Wald tests instead of the default score tests. The WALD option has no effect on the tests in the GEE parameter estimates table which are always Z statistics, which are just the square root of the Wald statistic and has the same p-value. The table that gives the Pearson and deviance statistics does not apply to GEE models since it is not a likelihood-based method. The table is printed if you specify the PRINTMLE option in the REPEATED statement, but those statistics apply only to an initial model that is fitted by maximum likelihood to provide starting values for the GEE estimation method. The GEE method always includes a dispersion parameter that is used in the computation of the working correlation matrix. See "Generalized Estimating Equations" in the Details section of the GENMOD documentation.