Programming the statistical procedures from SAS

Interpreting PROC Mixed Influence Diagnostics

Reply
Learner
Posts: 1

Interpreting PROC Mixed Influence Diagnostics

I'm a business user focused on interpretation of results (so I can muck around with somebody else's code and have enough of a statistical background to be dangerous).  I am looking for help in interpretting the outputs of PROC MIXED Influence Diagnostics.  

 

1.  My approach initially is to look at Cook's D and use a cutoff of 4/n, DFFITS with a cutoff of 2SQRT(p/n)  and examine if RMSE changes much for levels with or without this level.  is that a valid approach?

 

2.  I read breifly somewhere that I could use PRESS to examine overfitting - does anybody have any direction they coud provide on how I should thinkg aobut this?

 

3.  How should I interpret the other measures provided?

 

4.  I have a model that provides no diagnostics, providing notes only of Loss of Rank or Change in Singularity.  Need help in interpetting this (but I fear this means that my model is fragile and cannot live without all the levels that I'm using.  

 

5.  Anything that I'm missing / not thinking of?

 

Help greatly appreciated

Valued Guide
Valued Guide
Posts: 684

Re: Interpreting PROC Mixed Influence Diagnostics

In addition to the examples in the MIXED chapter of the User's Guide (available on-line), I recommend that you read the following:

http://www2.sas.com/proceedings/sugi29/189-29.pdf

 

 

Ask a Question
Discussion stats
  • 1 reply
  • 168 views
  • 2 likes
  • 2 in conversation