[R] Influence measure + lme ?

Spencer Graves spencer.graves at pdf.com
Sat Jan 21 17:51:35 CET 2006


	  My search just now found no mention of Cook's distance or influenc 
measures.  The closest I found was an unanswered question on this from 
April 2003 (http://finzi.psych.upenn.edu/R/Rhelp02a/archive/4797.html).

	  Beyond that, there is an excellent discussion of "Examining a Fitted 
Model" in Sec. 4.3 (pp. 174-197) of Pinheiro and Bates (2000) 
Mixed-Effects Models in S and S-Plus (Springer).  Pinheiro and Bates 
decided NOT to include plots of Cook's distance among the many 
diagnostics they did provide.  However, 'plot(fit.lme)' plots 
'standardized residuals' vs. predicted or 'fitted values'.  Wouldn't 
points with large influence stand apart from the crowd in terms of 
'fitted value'?

	  Of course, there are many things other one could do to get at related 
information, including reading the code for 'influence' and 'lme', and 
figure out from that how to write an 'influence' method for an 'lme' 
object.  However, that is far beyond what you asked and what I have time 
now to disucss.

	  Hope this helps.
	  spencer graves
	
Chia, Yen Lin wrote:

> Hi all,
> 
>  
> 
> Does lme has function to compute the cook's distance or influence
> measure like lm?  I can't find them.  Thanks.  
> 
>  
> 
> Yen Lin
> 
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html




More information about the R-help mailing list