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
- /
- SAS Programming
- /
- General Programming
- /
- How to handle rare events and imbalanced samples

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

09-23-2011 08:19 AM

I am developing a prediction model with a logistic regression by using SASEnterprise Miner. The original sample (N=342) only has 16 target “1” category,which corresponds to 4,7% (16/342) of the observations. To handle theimbalanced sample and the rare events issue, at the sample general propertypanel, for the level based option I set sample proportion as 50.0. Hence I endup with a sample containing 32 observations (16 observations for target “1”category and 16 observations for the target “0” category), which I used todeveloped the prediction model.

However my PhD Adviser is concern because “the occurrence probability from the scoring is not the one that predicts correctly for the originaldistribution, it is the one that predicts correctly for the oversampled set”.For my PhD thesis stand point of view, I must provide references that thismethodology “oversampling” approach is legible.

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

09-26-2011 09:46 PM

Frankly, I'd be worried about the stability of any logistic regression model that had only 16 outcomes. Harrell states that one needs 10-20 outcomes per candidate predictor variable (e.g. degree of freedom).

[ http://biostat.mc.vanderbilt.edu/wiki/Main/RmS#REGRESSION_MODELING_STRATEGIES ]

With that rule of thumb, you could only consider one variable.

Doc Muhlbaier

Duke