Re: [R] help with syntax of nlme function

From: Spencer Graves <spencer.graves_at_pdf.com>
Date: Mon 05 Jun 2006 - 06:02:26 EST

          I haven't seen a reply to this, so I will offer a few comments. I'm not sufficiently familiar with nlme to answer your question directly, but I hope my comments might help.

          First, have you studied Pinheiro and Bates (2000) Mixed-Effects Models in S and S-Plus (Springer)? If no, I highly recommend this book for you. It has three chapters with many examples on nonlinear mixed-effects models. The nlme package is the product of Doug Bates and many of his graduate students, including Jose Pinheiro. For over a quarter century, Doug Bates has been a leading, original contributor to the theory and practice of nonlinear and mixed-effects modeling. I was unable to get much out of lme (let alone nlme) until I read this book. If you are not familiar with this book, I believe it will introduce you to many "helper" functions that will make it easier for you to produce plots as well as p-values, etc., for your model. This in turn should help you get more information from your data and be able to describe your results more persuasively and interestingly to your audience.

          Also, can you check the directory on your computer where R is installed? It should have a subdirectory "library" that includes a further subfolder "nlme". The "script" folder in "nlme" contains script files containing R commands to produce most of the computer output in the book. Ch. 7 is a lot of math that you may wish to skip. However, if you work through the script files while reading chapters 6 and 8, you will hopefully find examples that are sufficiently close to your current application that you will be able to find and fix the problem.

          If that fails, try to find a simple, self-contained reproducible example that also displays your problem. With luck, you should be able to generate such by modifying one of the examples in the book, by reducing the size of the data set used or misspelling a name or something. Then submit another post containing this example.

          I'm sorry I can't give you a quicker answer.

	  Hope this helps,
	  Spencer Graves

Bill Shipley wrote:
> I am having difficulty understanding the syntax of the nlme() function for
> nonlinear mixed models. The data frame is called Marouane.chlorophyll. The
> model involves a dependent variable (Absorb) and an independent variable
> (Ch.surf), which are both numeric variables in the data frame. The data are
> hierarchically grouped as Espece, Plante nested within Espece, and Feuille
> nested within Plante (i.e. leaves within plants within species). Espece is
> defined as a factor and Plante and Feuille are defined as numeric in the
> data frame. There are two parameters (k1 and b) and I want to fit a model
> in which these two parameters vary randomly across Espece and Plante (with
> Feuille being the residual level). Here is how I have specified the call to
> nlme:
>

>> fit<-nlme(model=Absorb~ k1 - (1/(b*Ch.surf)),

> + data=Marouane.chlorophyll,fixed=k1+b~1,
> + groups=~Espece/Plante,
> + start=list(k1=97.8, b=4.68))
>
> However, there is something wrong with this syntax because I get the
> following error:
>
> Problem in names<-: Invalid length for names attribute: structu
> re(.Data = list(structure(.Data = numeric(0), class ....
>
> Can someone explain what I am doing wrong?
>
>
>
> Bill Shipley
>
> North American Editor, Annals of Botany
>
> Editor, "Population and Community Biology" series, Springer Publishing
>
> Département de biologie, Université de Sherbrooke,
>
> Sherbrooke (Québec) J1K 2R1 CANADA
>
> Bill.Shipley@USherbrooke.ca
>
> http://callisto.si.usherb.ca:8080/bshipley/
>
>
>
>
>
> [[alternative HTML version deleted]]
>
>
>
> ------------------------------------------------------------------------
>
> ______________________________________________
> R-help@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


R-help@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 Received on Mon Jun 05 06:10:03 2006

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.1.8, at Mon 05 Jun 2006 - 08:10:41 EST.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.