I am trying to introduce AR1 term into my models (lag of dependent variable). It is easy to do for the for the historic period which is lag(y). But I am wondering how to create it for the forecast period (it would actually be the lag of the prediction). Is ther any SAS procedure or autoreg can do this?? My data looks like:
Date Y lag(y) X1 X2
… .006 …. 12 7
Q1 2016 .005 .006 1 3
Q2 2016 .004 .005 11 7
Q3 2016 .003 .004 8 7
Q4 2016 .004 .003 10 6
Q1 2017 . .004 12 5
Q2 2017 . . 11 4
Q3 2017 . . 10 3
Q4 2017 . . 11 4
….
How do I populate lag(Y) for qtr’s Q2 2017 & beyond so that I can use it in regression??
Thanks in advance.
Input Variables and Regression with ARMA Errors
You can use PROC ARIMA from SAS/ETS with p = 1 (AR1) along with x1 and x2 as input variables.
proc arima data=<yourdata>; identify var=y crosscorr=(x1 x2); estimate p=1 input=(x1 x2); run;
Input Variables and Regression with ARMA Errors
You can use PROC ARIMA from SAS/ETS with p = 1 (AR1) along with x1 and x2 as input variables.
proc arima data=<yourdata>; identify var=y crosscorr=(x1 x2); estimate p=1 input=(x1 x2); run;
You want a one period lead for Y. I like @alexchien's answer, but here is a way to get a one period lead:
data need;
merge have
have (firstobs=2 keep=y rename=(y=y_lead1));
run;
Don't miss out on SAS Innovate - Register now for the FREE Livestream!
Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.
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.