Re: [R-sig-phylo] fitting multiple OU optima

2014-04-24 Thread Carl Boettiger
Hi Paolo, Though bear in mind that making such comparisons by AIC can be rather misleading, as we illustrated here http://dx.doi.org/10./j.1558-5646.2011.01574.x , and a variety of alternative exist. (With apologies for highlighting my own work), - Carl On Tue, Apr 22, 2014 at 5:46 AM,

Re: [R-sig-phylo] fitting multiple OU optima

2014-04-22 Thread Florian Boucher
Hi Paolo, the method for doing what you want is called 'SURFACE' and there is an associated R package: http://cran.r-project.org/web/packages/surface/index.html The paper describing the method can be found here: http://onlinelibrary.wiley.com/doi/10./2041-210X.12034/abstract Cheers, Florian

[R-sig-phylo] fitting multiple OU optima

2014-04-22 Thread Santiago Sanchez-Ramirez
Hi Paolo, I'm not sure if what you ask has been implemented yet in any package, but something that you can do, in the case it hasn't, is to set up models with as many regimes as you consider realistic and feasible and then compare them using AIC. Cheers, Santiago