Predictive regression on next year's independent variables

[wryd078]

Hi SAS Community,

I'm currently using a macro to run rolling regressions where every year the data size increases, i.e. 2009, 2009-2010, 2009-2011 etc., with the following code. 

ods graphics off;

ods listing close;

%macro loopOverYears;

%do i=2009 %to 2018;

ods output ParameterEstimates=Output&i;

proc sort data=import; by GIC_SECT Year;

proc glm data=import;

class GIC_SECT Year;

model logtotal = GIC_SECT Year logrev logassets logcapx gmar levg / solution;

where year<=&i;

run;

%if &i = 2009 %then %do;

data AllEstimates;

set output&i;

endYear=&i;

run;

%end;

%else %do;

data output&i;

set output&i;

endYear=&i;

run;

data AllEstimates;

set AllEstimates output&i;

run;

%end;

%end;

ods listing;

%mend;

%loopOverYears;

From here, I need to be able to use the coefficients from these regression results to predict next year's logtotal based on next year's independent variables (GIC_SECT Year logrev logassets logcapx gmar levg). For example, I would run a regression for a data subset 2009-2015 and then use these results to predict logtotal in 2016 based on the independent variables in 2016. 

Does anyone know how I might accomplish this? 

[sbxkoenk]

These three articles should help you out. They give an exhaustive answer to your scoring question.

1.

Techniques for scoring a regression model in SAS

By Rick Wicklin on The DO Loop  February 19, 2014

https://blogs.sas.com/content/iml/2014/02/19/scoring-a-regression-model-in-sas.html

2.

Paper SAS3337-2019

SCORE! Techniques for Scoring Predictive Regression Models Using SAS/STAT® Software

Phil Gibbs and Randy Tobias, SAS Institute Inc.

https://www.sas.com/content/dam/SAS/support/en/sas-global-forum-proceedings/2019/3337-2019.pdf

3.

Usage Note 33307: Scoring (computing predicted values) for new observations or a validation data set

http://support.sas.com/kb/33/307.html

As you will clearly notice, you have multiple options to achieve what you are after.

Koen