Programming the statistical procedures from SAS

Second order autoregressive covariance structure in multilevel model

Accepted Solution Solved
Reply
Occasional Contributor
Posts: 8
Accepted Solution

Second order autoregressive covariance structure in multilevel model

Does SAS have the native ability to implement a second order autoregressive covariance structure within either PROC MIXED or PROC GLIMMIX?  I am aware that I can specify a first order autoregressive structure with TYPE = AR(1) on either the RANDOM or REPEATED statement line.  However, using AR(2) returns and error of "Invalid order specification".

I am surprised that AR(2) does not appear to be a valid function.

Thanks,

Matt


Accepted Solutions
Solution
‎04-30-2014 12:54 PM
Respected Advisor
Posts: 2,655

Re: Second order autoregressive covariance structure in multilevel model

Not available.  Do you think you could approximate it with either ANTE(1) or TOEPH?  Those look like they could be AR(2) under the right parameterization.  However, this really falls into the wild a&&ed guess category.

Steve Denham

View solution in original post


All Replies
Solution
‎04-30-2014 12:54 PM
Respected Advisor
Posts: 2,655

Re: Second order autoregressive covariance structure in multilevel model

Not available.  Do you think you could approximate it with either ANTE(1) or TOEPH?  Those look like they could be AR(2) under the right parameterization.  However, this really falls into the wild a&&ed guess category.

Steve Denham

Occasional Contributor
Posts: 8

Re: Second order autoregressive covariance structure in multilevel model

Thanks, Steve.  I appreciate you confirming my suspicions. 

🔒 This topic is solved and locked.

Need further help from the community? Please ask a new question.

Discussion stats
  • 2 replies
  • 306 views
  • 3 likes
  • 2 in conversation