topic Re: How to create a predictive model using test and train datasets available in Statistical Procedures

How to create a predictive model using test and train datasets available

Vish33 — Thu, 11 May 2017 13:10:23 GMT

Hello all,

I want to create a basic model which would identify/predict customers those are eligible to get the loan so that these customers would be targeted by the customer care people.

Please find the attachments for sample documents. I am new to modelling , so i would like to know how this can be done in SAS.

And i want to automate this finally to run on weekly/monthly basis.

Thank you.

Vish

Re: How to create a predictive model using test and train datasets available

StatDave — Thu, 11 May 2017 14:29:17 GMT

Assuming that your response is binary (eligible vs not), you would be fitting a logistic model. You can use PROC LOGISTIC to fit the model and the SCORE statement in that procedure to get predicted probabilities of eligibility for you test data set. See the LOGISTIC documentation and examples there such as the "Scoring data sets" example. Basic code would be like:

proc logistic data=train;

model binary_response(event="Y") = predictor1 predictor2;

score data=test out=mypreds;

run;

Re: How to create a predictive model using test and train datasets available

Vish33 — Thu, 11 May 2017 14:44:06 GMT

Thank you for your inputs. I will look into the Logistic documentation.

Re: How to create a predictive model using test and train datasets available

Ksharp — Fri, 12 May 2017 13:19:25 GMT

Also you could try Decision Tree.

Check PROC HPSPLIT example .