I am trying to forecast month end balances of the TOTAL portfolio using PROC ARIMA procedure. INside the portfolio there are 3 major product segments A, B and C. Each of which has a separate trend and seasonality as shown below in the graph
I wanted to try two methods,
method 1: Forecast of the TOTAL portfolio balance using PROC ARIMA (univariate)
method 2: Forecast of the TOTAL portfolio balance using PROC ARIMA (multi variate) using segment A, B and C BALANCES as input variables
Need help in forecasting the portoflio balance with multiple input variables, like average term , average interest rate, prepayments etc..
Which procedure can I use to achieve this?
Method 1 works fine but with method 2 I am getting these warnings
More values of input variable A are needed
More values of input variable B are needed
More values of input variable C are needed
METHOD 1:
proc arima data=upb_2 plots=all;
identify var=UPB(1,12) noprint;
estimate noconstant method=ml;
forecast lead=17 interval=month id=RPT_PRD_DATE out=outf;
run;
METHOD 2:
proc arima data=UPB_INDUSTRY_2 plots=all;
identify var=TOTAL(1,12) noprint crosscorr=(A B C);
estimate input=(A B C) noconstant method=ml;
forecast lead=17 interval=month id=RPT_PRD_DATE_2 out=outf_input;
run;
Hi @sivakoya ,
Method 2 in your example is an ARIMA model with input variables. To forecast a response series using an ARIMA model with inputs, PROC ARIMA needs values of each of the input series for the LEAD= forecast horizon. You can either provide future values of the input variables in the DATA= data set (setting the response variable to missing for those future observations) or you can use PROC ARIMA to model and forecast the input variables in addition to the response series.
For more details, please see the Forecasting with Input Variables section of the PROC ARIMA documentation at the following link:
I hope this helps!
DW
Hi @sivakoya ,
Method 2 in your example is an ARIMA model with input variables. To forecast a response series using an ARIMA model with inputs, PROC ARIMA needs values of each of the input series for the LEAD= forecast horizon. You can either provide future values of the input variables in the DATA= data set (setting the response variable to missing for those future observations) or you can use PROC ARIMA to model and forecast the input variables in addition to the response series.
For more details, please see the Forecasting with Input Variables section of the PROC ARIMA documentation at the following link:
I hope this helps!
DW
Thanks
Registration is now open for SAS Innovate 2025 , our biggest and most exciting global event of the year! Join us in Orlando, FL, May 6-9.
Sign up by Dec. 31 to get the 2024 rate of just $495.
Register now!
Learn how to run multiple linear regression models with and without interactions, presented by SAS user Alex Chaplin.
Find more tutorials on the SAS Users YouTube channel.