BookmarkSubscribeRSS Feed
jaweriahh
Obsidian | Level 7
While estimating an Unobseved Components Model how can i get the prediction error variance in my output?
3 REPLIES 3
Puwang
Obsidian | Level 7

You can use the OUTFOR= option in the FORECAST statement. The table contains the "STD" column which stores the standard error of prediction. Is that what you want?

 

You can also refer to the online help doc of UCM procedure:

http://support.sas.com/documentation/cdl/en/etsug/60372/HTML/default/viewer.htm#etsug_ucm_sect036.ht...

jaweriahh
Obsidian | Level 7

 

Thank you for your reply. I require some common diagnostic statistics such as (Predicted error variance

the one step ahead prediction errors, the normality statistic based on the third and fourth moments, heteroscedasticity statistic based on the ratio of the sample for the first one third of the prediction errors, the Box-Ljung serial correlation statistics, Durbin Watson) I have looked into the outfor=option but could not get these statistics (though I was able to get several diagnostic graphs).

Think you

Puwang
Obsidian | Level 7

Use ODS table options to trieve more statistics. For example, the following statements can provide you the one step ahead prediction errors, the normality statistic.

   ods output FitStatistics = fitstat;
   ods output FitSummary = fits;

 

More info on ODS output tables please check

http://support.sas.com/documentation/cdl/en/etsug/60372/HTML/default/viewer.htm#etsug_ucm_sect034.ht...

sas-innovate-white.png

Our biggest data and AI event of the year.

Don’t miss the livestream kicking off May 7. It’s free. It’s easy. And it’s the best seat in the house.

Join us virtually with our complimentary SAS Innovate Digital Pass. Watch live or on-demand in multiple languages, with translations available to help you get the most out of every session.

 

Register now!

Discussion stats
  • 3 replies
  • 2080 views
  • 0 likes
  • 2 in conversation