On Wed, Feb 9, 2011 at 1:02 PM, Anthony Lawrence Nguy-Robertson <anthony.robert...@huskers.unl.edu> wrote: > Thank you R-forum for you generous help. > > Gabor Grothendieck, I am not sure if anova in the form that you suggested is > the most appropriate (This is probably more of a statistics related, rather > than R related at this point). The way I understand anova is that you are > testing the variance between the models. I know that the variance in the > 'corn' models is greater than 'soybean' due to a biological reason. I don't > think this approach is correct in my case since the anova analysis is > comparing variance between different data sets.
Yes, you do an F test between the models using anova.nls except the models I specified in my response are not the models that you used and I think you are misinterpreting my answer even though I did explicitly point out that the models I was referring to were not the models in your post. The models you used can't be compared this way since they each refer to a different data set. I corrected the approach in your question by using nested models which both refer to the same combined data set. "Just to be clear the two models would each include both groups -- one model would assume the parameters are the same for the two groups so it would have 3 parameters and the other model would allow them to be different (up to 6 parameters depending on how many parameters you wish to allow to be different between the two groups) -- these are not the models shown above." -- Statistics & Software Consulting GKX Group, GKX Associates Inc. tel: 1-877-GKX-GROUP email: ggrothendieck at gmail.com ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
