BookmarkSubscribeRSS Feed
itslarajean
Fluorite | Level 6
Hi all,
I have a VAR regression produced out based on my Development data period. I now wish for this VAR model to predict the future values but with some variables to have a fixed value (that variable will not be forecasted by my VAR model).

For example:
My VAR model is below
A = a1A_t-1 + b1B_t-1 + c1C_t-1
B = a2A_t-1 + b2B_t-1 + c2C_t-1
C = a3A_t-1 + b3B_t-1 + c3C_t-1

I want the VAR model to predict A and B but for C I have a set of values I wish to input. How do I do this?

Thank you.
1 REPLY 1
sbxkoenk
SAS Super FREQ

I have moved this topic to the 'SAS Forecasting and Econometrics' board.

 

Hello @itslarajean ,

 

Are you using PROC VARMAX?

 

I think you can use "the missing value trick for scoring a vector ARIMA(X) model".
As described in the below blog for regression ...

The missing value trick for scoring a regression model
By Rick Wicklin on The DO Loop February 17, 2014
https://blogs.sas.com/content/iml/2014/02/17/the-missing-value-trick-for-scoring-a-regression-model....

 

Include training (development as you call it) and future periods in your dataset (the one that you put in the DATA= option on the PROC VARMAX statement).

For future periods, A and B are missing , but C has values.

 

I hope it works for you,

Koen

sas-innovate-2026-white.png



April 27 – 30 | Gaylord Texan | Grapevine, Texas

Registration is open

Walk in ready to learn. Walk out ready to deliver. This is the data and AI conference you can't afford to miss.
Register now and save with the early bird rate—just $795!

Register now

Discussion stats
  • 1 reply
  • 1162 views
  • 0 likes
  • 2 in conversation