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
- /
- Logistic regression question(s)

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

01-24-2013 01:45 PM

Greetings all. I am wanting to run logistic regression on some data, and it has been a while since I've done any stats (and I learned on SPSS). I have a binary response variable, two predictor variable of interval type, and several binary dummy variable predictors. I'm not sure how to go about running the analysis. I have messed around with both catmod and genmod, but I'm not sure if I'm doing it right. Basically, I'm looking for significance of the predictors in the response "chg_off". I've attached a sas file with some sample data. I took a random sample of 100 observations, so there may be nothing to the results, I'm more just looking for hot to set it up. I sure would appreciate any help anyone could offer. Thank you.

Greg

Accepted Solutions

Solution

01-24-2013
01:48 PM

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

01-24-2013 01:48 PM

The SAS Examples in proc logistic are pretty good IMO.

All Replies

Solution

01-24-2013
01:48 PM

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

01-24-2013 01:48 PM

The SAS Examples in proc logistic are pretty good IMO.

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

01-29-2013 09:56 AM

Thank you Reeza, that is just what I needed. I spent some time googling "sas logistic regression", and I came up with genmod, catmod, but not logistic. Anyhow, thanks for the tip.

Greg

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

01-29-2013 04:54 PM

Ok, one more question regarging this. I am looking at the "Analysis of Maximum Likelihood Estimates" output, and I am confused as to why some of the variables have '<.0001' in the 'Pr > ChiSq' column. I would (maybe naively) infer these variables to be significant. Am I wrong in this assumption? It just seems to straight forward. Thank you.

Greg

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

01-29-2013 05:13 PM

That means P-Value is less than 0.0001, it just won't display it smaller....