BookmarkSubscribeRSS Feed
Jane7476
Calcite | Level 5

Hi,

My question is will 'archtest, reset and godfrey' tests produce different output for a model when

- you include the weights option in proc reg

- exclude the weight option.

Alternative that I have is that:

I need these 3 tests but my problem is that proc reg doesn't support these tests as default code options. Autoreg supports these options but I don't think it has a weights option. So is there any way of producing these three tests in proc reg?

I am running following code:

proc reg data=test;

model actual=ind1 ind2;

restrict ind1=xxx;

weight w1;

quit;

proc autoreg data=test;

model actual=ind1 ind2/archtest reset godfrey;

quit;

Any thoughts?

Regards,

1 REPLY 1
SteveDenham
Jade | Level 19

Short answer is no, PROC REG will not do this with any standard options.

Long answer--there may be a way to 'weight' observations and get the Godfrey, archtest and reset tests if you shift to PROC MODEL.  It will take some coding, and I am afraid I am not the right person for that.  Consider reposting this to the Forecasting and Econometrics forum--either Ken or Udo may have some ideas.

Steve Denhamn

sas-innovate-white.png

Missed SAS Innovate in Orlando?

Catch the best of SAS Innovate 2025 — anytime, anywhere. Stream powerful keynotes, real-world demos, and game-changing insights from the world’s leading data and AI minds.

 

Register now

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 1 reply
  • 1702 views
  • 0 likes
  • 2 in conversation