[R-sig-ME] lack of convergence

[Highland Statistics Ltd.]

> I already ran a mixed model and got convergence.  I have 700 data points.  Now, I am running a new mixed model that adds the ARMA (corARMA) covariance pattern.  I keep getting an error message telling me that the model did not reach convergence.  I already tried reducing my tolerance to 1e-1 and increased the number of iterations to 7,000.  Still, I am not having any luck.
>
> Any suggestions to get my model to work properly so I can get my parameter estimates?  I pasted my covariance code below.
>
> LVB2.nlme<-update(LVB.nlme, corr=corARMA(p=1, q=1),
> control=nlmeControl(maxIter=6000, tolerance=1e-1, msVerbose=TRUE))
>


Mike,

Specifying starting values for the AR and MA parameters inside the 
corARMA function will help. You can use the ACF to choose decent 
starting values.

Keep in mind that the correlation is imposed inside your random effect.

Alain

------------



> _______________________________________________
> R-sig-mixed-models mailing list
> R-sig-mixed-models at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
>
> End of R-sig-mixed-models Digest, Vol 55, Issue 11
> **************************************************
>


-- 


Dr. Alain F. Zuur
First author of:

1. Analysing Ecological Data (2007).
Zuur, AF, Ieno, EN and Smith, GM. Springer. 680 p.
URL: www.springer.com/0-387-45967-7


2. Mixed effects models and extensions in ecology with R. (2009).
Zuur, AF, Ieno, EN, Walker, N, Saveliev, AA, and Smith, GM. Springer.
http://www.springer.com/life+sci/ecology/book/978-0-387-87457-9


3. A Beginner's Guide to R (2009).
Zuur, AF, Ieno, EN, Meesters, EHWG. Springer
http://www.springer.com/statistics/computational/book/978-0-387-93836-3


Other books: http://www.highstat.com/books.htm


Statistical consultancy, courses, data analysis and software
Highland Statistics Ltd.
6 Laverock road
UK - AB41 6FN Newburgh
Tel: 0044 1358 788177
Email: highstat at highstat.com
URL: www.highstat.com
URL: www.brodgar.com