[R] AR(1) with NLME

Spencer Graves spencer.graves at pdf.com
Sat Nov 5 03:02:46 CET 2005


	  Have you plotted the residuals (without the AR(1))?  Lack of fit can 
masquerade as highly autocorrelated residuals.

	  Alternatively, can you still estimate your model using the first 
differences of your data?  If yes, do you have a problem?

	  Have you studied the appropriate portions of Pinheiro and Bates 
(2000) Mixed-Effects Models in S and S-PLUS (Springer)?  If you'd stiill 
like help from this list, PLEASE do read the posting guide! 
"www.R-project.org/posting-guide.html", especially the part about 
providing a toy example in a few lines of code that someone can copy 
from your email into R, try a few things, and craft a response in a 
minute or two.  Doing that will likely increase the speed and utility of 
replies.

	  Best Wishes,
	  Spencer Graves

Robert Schneider wrote:

> I am trying to calibrate a non linear mixed model with a AR(1)
> autocorrelation structure. This structure is indicated by the ACF plot. When
> calibrating the model without the AR(1), everything comes out nicely.
> However, when the AR(1) structure is included in the model, the Phi
> parameter estimate comes out to be 1. This result occurs with various
> starting values in the correlation statement. Am I missing something, or the
> model cannot be calibrated with the AR(1)?  
> 
>  
> 
> Thanks for you input.
> 
>  
> 
> Robert Schneider
> 
>  
> 
>  
> 
> 
> 	[[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

-- 
Spencer Graves, PhD
Senior Development Engineer
PDF Solutions, Inc.
333 West San Carlos Street Suite 700
San Jose, CA 95110, USA

spencer.graves at pdf.com
www.pdf.com <http://www.pdf.com>
Tel:  408-938-4420
Fax: 408-280-7915




More information about the R-help mailing list