applying multinomial logistic model to new data

[Tanya Dockendorf-2]

Hi everyone,

I need to do something in a hurry that I don’t normally do, so I hope someone has an easy answer. I ran a multinomial logistic regression on a set of data and need to apply the model to classify cases from another data set. In the new data, I have all independent variables that were used in building the regression model, but not the dependent variable, which is why I need to apply the model. Is there a quick way to do that in spss? Typically I would program this in Excel or another software, but in this case, that would take too long.

 

Thanks in advance!

 

Tanya

 

 

[Bruce Weaver]

Tanya Dockendorf-2 wrote

Hi everyone,
I need to do something in a hurry that I don't normally do, so I hope someone has an easy answer. I ran a multinomial logistic regression on a set of data and need to apply the model to classify cases from another data set. In the new data, I have all independent variables that were used in building the regression model, but not the dependent variable, which is why I need to apply the model. Is there a quick way to do that in spss? Typically I would program this in Excel or another software, but in this case, that would take too long.

Thanks in advance!

Tanya

Merge the two data sets with ADD FILES, then run your model, and have it save whichever of the following variables you want:

Saved variables for NOMREG:

• Estimated response probabilities. These are the estimated probabilities of classifying a factor/covariate pattern into the response categories. There are as many estimated probabilities as there are categories of the response variable; up to 25 will be saved.

• Predicted category. This is the response category with the largest expected probability for a factor/covariate pattern.

• Predicted category probabilities. This is the maximum of the estimated response probabilities.

Because the dependent variable does not exist in the second dataset, those data will not be used in generating the model.  But the saved variables will appear for every record that has all of the explanatory variables.

HTH.

[Tanya Dockendorf-2]

Thank you! This solution worked out great. I didn't realize that NOMREG will write out probabilities / predicted category for the whole data set in this case.

Thanks again,
Tanya


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Bruce Weaver
Sent: Wednesday, July 14, 2010 2:01 PM
To: [hidden email]
Subject: Re: applying multinomial logistic model to new data

Tanya Dockendorf-2 wrote:

>
> Hi everyone,
> I need to do something in a hurry that I don't normally do, so I hope
> someone has an easy answer. I ran a multinomial logistic regression on a
> set of data and need to apply the model to classify cases from another
> data set. In the new data, I have all independent variables that were used
> in building the regression model, but not the dependent variable, which is
> why I need to apply the model. Is there a quick way to do that in spss?
> Typically I would program this in Excel or another software, but in this
> case, that would take too long.
>
> Thanks in advance!
>
> Tanya
>

Merge the two data sets with ADD FILES, then run your model, and have it
save whichever of the following variables you want:

Saved variables for NOMREG:

* Estimated response probabilities. These are the estimated probabilities of
classifying a factor/covariate pattern into the response categories. There
are as many estimated probabilities as there are categories of the response
variable; up to 25 will be saved.

* Predicted category. This is the response category with the largest
expected probability for a factor/covariate pattern.

* Predicted category probabilities. This is the maximum of the estimated
response probabilities.

Because the dependent variable does not exist in the second dataset, those
data will not be used in generating the model.  But the saved variables will
appear for every record that has all of the explanatory variables.

HTH.


-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/applying-multinomial-logistic-model-to-new-data-tp1092898p1092962.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD