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
- /
- Offsetting in proc genmod

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
- Highlight
- Email to a Friend
- Report Inappropriate Content

02-17-2012 02:21 PM

Hello-

Could someone help me with the following question?

I tried to offset several factors at the same time in a GLM model using SAS. There is an Offset option in Proc Genmod but it seems that you can only use it to offet a factor per time. Dose SAS provide other option or statement that can be used to offset several factors simultaneously?

Thanks advance.

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

02-20-2012 07:52 AM

The offset option essentially scales the response variable. I think you are asking about scaling independent variables. Is that correct? If so, you will probably have to apply a dataset transformation, or include protgramming statements within PROC GENMOD.

Steve Denham

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

02-20-2012 08:50 AM

Thank you very much, Steve. You are right. I am going to scale independent variables. Would you please specify the way to transform the dataset or what kind of protgramming statement I should use? Really appreciated for your help.

Meta

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

02-21-2012 07:19 AM

To answer that is as much art as science, I am afraid. You need to look at the processes that generate the independent variables--what sort of distribution results, are there collinearities or redundancies, are the variables straightforward so that a parametric distribution is appropriate or would a semiparametric (B spline) be better?

A first step might be to look at the residuals after standardizing the independent variables--and I do mean LOOK. A graphic presentation will tell you far more than anything else. Plot, plot, and re-plot the data, so that you get a feel for what can be done.

I am afraid I haven't given much of an answer, but then I don't really have a good idea of what your variables might be.

Steve Denham

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

02-21-2012 09:24 AM

Thank you very much, Steve.

I might misunderstand your first suggestion and I did not clearly describe my question. The offsetting I mean here is to "fix" the relativities of independent variables either because they are known or they are selected due to business competition. Sorry for confusion. Really appreciated for your help. If you know some solutions, would you please give me some suggestions? Thanks again,

Meta

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

02-22-2012 08:32 AM

Since we are way outside of my field (biostatistics), there may be some terminology problems. What do you mean when you say "fix the relativities of independent variables"? If I knew or saw an example, then I might be able to come up with an analogous situation in my area.

Steve Denham

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

02-22-2012 09:18 AM

Thank you very much, Steve.

For example, In bank A the model found that female card-holder is relatively riskless or fewer chance to pay later than due date compared to male cardholder and the model relativity is 1: 1.15 (ie., male is 15% higher chance to have late payment). Let us say, the model suggests to charge male cardholder 15% higher than female. But the bank decided to charge male only 10% higher due to competition or other business consideration. If the Gender factor is correlated to other factors in the model, the fixing (offsetting) relativity to 1:1.1 might affect the estimated parameters of other factors. Using offsetting the Gender relativity, we can see what is the effect by applying this business decision.

If you can found a way to do the offseting job in proc genmod, would you please let me know.

Really appreciated for your help.

Meta