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
- /
- SAS Procedures
- /
- Out of sample predictions

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

09-15-2010 11:54 AM

Once I've estimated a regression model, how do I use that model to make out of sample predictions for y-hat. I have a large number of MSA dummy variables (202), so creating a new variable that manually calculates that value would be very tedious and vulnerable to human error.

Any help would be greatly appreciated.

Thanks,

Jim

Any help would be greatly appreciated.

Thanks,

Jim

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

Posted in reply to deleted_user

09-15-2010 02:11 PM

You need to create a data set with the values of the independent variables that you want to make predictions for, concatenate this data with the original data set (leaving the value of the response variable as missing), then rerun the regression model specifying the P option in the model statement.

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

Posted in reply to polingjw

09-15-2010 03:42 PM

And after you follow polingjw's advice, keep your fingers crossed that the predicted values mean anything.

If you really have 202 dummy variables, I'm skeptical that any such prediction of out of sample values will have meaning. More than likely, you will be extrapolating, in other words, predicting in a region of X-space where you have no data.

If you really have 202 dummy variables, I'm skeptical that any such prediction of out of sample values will have meaning. More than likely, you will be extrapolating, in other words, predicting in a region of X-space where you have no data.

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

Posted in reply to Paige

09-15-2010 04:26 PM

Thanks for your help!

The model is an MSA fixed effects model (thus the 200 MSA dummies). I have 142,000 observations in my sample and 38,000 in the sub-sample where the model is estimated.

I have the y-values for the entire sample, I just want to see if the predicted values from the sub-sample model are close to the values in the entire sample.

Not sure if this changes anything.

The model is an MSA fixed effects model (thus the 200 MSA dummies). I have 142,000 observations in my sample and 38,000 in the sub-sample where the model is estimated.

I have the y-values for the entire sample, I just want to see if the predicted values from the sub-sample model are close to the values in the entire sample.

Not sure if this changes anything.

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

Posted in reply to deleted_user

09-15-2010 03:43 PM

I am still working on estimating a regression model...

Could you help me with my post problem?

Could you help me with my post problem?