[R-sig-eco] growth curves - grofit

[Derek Ogle]

Ana,

This sounds like the same setup for which fisheries scientists often use an "Age-Length Key."  Here is one resource with a short description of the problem and one algorithmic solution from the fisheries literature -- http://fishr.wordpress.com/vignettes/ (sixth bulleted link on that page ... some of the code in that document is out-of-date, if you use that method let me know and I can explain how to update it).

By the way, I am not familiar with grofit but I would suspect that it fits models with age as the independent variable, not the dependent variable that your question implies.

Good luck with your work.

-----Original Message-----
From: r-sig-ecology-bounces at r-project.org [mailto:r-sig-ecology-bounces at r-project.org] On Behalf Of Romine, Jason
Sent: Wednesday, August 13, 2014 1:13 PM
To: A.P.B. Carneiro
Cc: r-sig-ecology at r-project.org
Subject: Re: [R-sig-eco] growth curves - grofit

Hi Ana,

I would simply select the models you think may fit your data, then use nlm or optim to fit the models.  Once you have selected your final model, use your data to predict ages from wing lengths.

Best,
Jason


--------------------------------------------------------
Jason G. Romine, Ph.D.
USGS Western Fisheries Research Center
Columbia River Research Laboratory
5501A Cook-Underwood Road
Cook, WA 98605
Phone: (509) 538-2299 x213
Email:jromine at usgs.gov
Website: http://wfrc.usgs.gov
---------------------------------------------------------


On Wed, Aug 13, 2014 at 4:55 AM, A.P.B. Carneiro <apbc2 at cam.ac.uk> wrote:

> Dear all,
>
> I am trying to use grofit package for biological growth curves. I 
> would like to see which curves describe best my data and from them 
> predict results I don’t have.
>
> I basically have a sample of data containing wing length of some birds 
> (as a growth measure) and their respective ages. And I have another 
> sample just with wing lengths (not associated with an age). The format 
> is a little different than what is specified on the package. What I would like to do is:
>
> 1)      Find the best curve that describes my data (using the data that I
> have ages),
>
> 2)      Apply the results to the data I don’t have ages, and predict these
> ages.
>
>
> Is it possible with grofit? Any ideas how to do it?
>
> Thanks,
>
> Ana
>
> --
>
> PhD student
> Evolutionary Ecology Group
> Department of Zoology
> University of Cambridge
>
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
>

	[[alternative HTML version deleted]]

_______________________________________________
R-sig-ecology mailing list
R-sig-ecology at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology