BookmarkSubscribeRSS Feed
Obsidian | Level 7

Hello, I have an unusual time series:

The first time series has a seasonality of 24 and 168 (24x7). How would we specify this in Proc ARIMA ?

For instance if my assumed model is (1,1,2) X (1,1,2)24 X (2,2,1)168, how would I doe this in SAS proc arima ?

The second time series has a seasonality of 52 but within the 52 weekly we have 4 seasonal cycles (13 X 4 = 52), can you please let me know how I would specify these in Proc ARIMA ?


SAS Employee

Maybe this:

identify var=y(1,24,168,168); /*double differencing at lag 168??*/

estimate p=(1)(24)(168 336) q=(1 2)(24 48)(168);

I am not sure about the 13x4 case. It depends on your model. But you can difference the series at different lags as much as you like, and specify AR and MA coefficients. Or use seasonal dummy variables.

SAS Employee

Hello Forecaster,

Your problems appear a little involved.  There are a few different ways to approach this and besides ARIMA, we might be able to use UCM procedure for these problems.  Is it possible to take this discussion offline where you provide me additional details and possibly your data (if necessary suitably altered to keep it private).  I support ARIMA and UCM.  Once you get a satisfactory answer, we could post it again (or not) in this list.  Send me an email at




Registration is open! SAS is returning to Vegas for an AI and analytics experience like no other! Whether you're an executive, manager, end user or SAS partner, SAS Innovate is designed for everyone on your team. Register for just $495 by 12/31/2023.

If you are interested in speaking, there is still time to submit a session idea. More details are posted on the website. 

Register now!

Multiple Linear Regression in SAS

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.

Discussion stats
  • 2 replies
  • 3 in conversation