Solved
New Contributor
Posts: 2

# How to score (with 95% confidence values) each observation in GLM after developing the model

Hi, i have used Proc GLM to develop a model on modelling data set .

Now i want to use the model (or model equation) to score the observations in a new data set, along with calculating the uncertainty for each observation (i.e. 95 % confidence values Lower and Higher for each observation in the new data set).

But in my version of SAS (i believe it is 8.0), i dont have proc plm to use. Also can i use proc score to get the above mentioned values ??

Thanks

Accepted Solutions
Solution
‎01-04-2012 08:06 AM
Posts: 2,655

## Re: How to score (with 95% confidence values) each observation in GLM after developing the model

There is an interesting work-around.  Append the new data set to the old, but replace all of the response variable values in the new data set with missings " . " before appending.

The fit will use only the original data, but the output data set (you'll need an OUTPUT statement that looks something like:

OUTPUT OUT=predset p lcl ucl;

This should give predicted values for all records with complete data for the independent variables, as well as confidence bounds.

Steve Denham

Message was edited by: Steve Denham

All Replies
Solution
‎01-04-2012 08:06 AM
Posts: 2,655

## Re: How to score (with 95% confidence values) each observation in GLM after developing the model

There is an interesting work-around.  Append the new data set to the old, but replace all of the response variable values in the new data set with missings " . " before appending.

The fit will use only the original data, but the output data set (you'll need an OUTPUT statement that looks something like:

OUTPUT OUT=predset p lcl ucl;

This should give predicted values for all records with complete data for the independent variables, as well as confidence bounds.

Steve Denham

Message was edited by: Steve Denham

New Contributor
Posts: 2

## How to score (with 95% confidence values) each observation in GLM after developing the model

Thanks a lot Steve, Interestingly i figured this very way out by myself before reading your reply.......

Still, thanks a lot for the reply, nd yeah this is an interesting workaround

🔒 This topic is solved and locked.