Re: [R-sig-phylo] Interpretation of standard errors of parameter estimates in OUwie models

Dear R-sig-phylo contributors, Thanks a lot for sharing your illuminating thoughts on the list. I agree with points raised by others, and I'd like to add one suggestion. As already mentioned by several, alpha and sigma-squared can be hard to estimate with certainty. However, in some cases, the

Re: [R-sig-phylo] Interpretation of standard errors of parameter estimates in OUwie models

Aloha Rafael, I am very glad that you are thinking about power of your analyses and confidence in your parameter estimates. Folks have chimed in with many useful responses, however, I believe your original question was whether uncertainty in parameter estimates indicate bad models and limits

Re: [R-sig-phylo] Interpretation of standard errors of parameter estimates in OUwie models

Apparently the plot did not come through correctly. I'm attaching a PDF of it. Best, Brian ___ Brian O'Meara, http://www.brianomeara.info, especially Calendar , CV

Re: [R-sig-phylo] Interpretation of standard errors of parameter estimates in OUwie models

Hi, Rafael et al. ?OUwie has some information on the standard errors: The Hessian matrix is used as a means to estimate the approximate standard > errors of the model parameters and to assess whether they are the maximum > likelihood estimates. The variance-covariance matrix of the estimated >

Re: [R-sig-phylo] Interpretation of standard errors of parameter estimates in OUwie models

Dear Rafael. I believe the standard errors are computed from (negative inverse of) the Hessian matrix - which is a matrix containing the second-order partial derivatives (or some numerical approximation of them) from the likelihood surface. These values are measurements of the curvature of

Re: [R-sig-phylo] Interpretation of standard errors of parameter estimates in OUwie models

Even if those models may fit the data better, they may not necessarily help Rafael determine whether or not the parameter estimates of interest are different across regimes (though perhaps BMS might be informative). Rafael, maybe you could try fixing the ancestral regimes to match most likely

Re: [R-sig-phylo] Interpretation of standard errors of parameter estimates in OUwie models

Yup, I ran all models available in OUwie and support generally follows model complexity. From greater to lesser support: OUMVA, OUMV, OUM, OU1, BMS, BM1. *--* *Rafael Sobral Marcondes* PhD Candidate (Systematics, Ecology and Evolution/Ornithology) Museum of Natural Science

Re: [R-sig-phylo] Interpretation of standard errors of parameter estimates in OUwie models

Hi Rafael, Have you tried running the BM1, BMS, or OU1 models? If so, how do the AICc values for those compare to those of the OUM model? If not, you should make sure you run those. If you find that the these models fit your data better, this could indicate that there isn't a large different

[R-sig-phylo] Interpretation of standard errors of parameter estimates in OUwie models

Dear all, I'm writing (again!) to ask for help interpreting standard errors of parameter estimates in OUwie models. I'm using OUwie to examine how the evolution of bird plumage color varies across habitat types (my selective regimes) in a tree of 229 tips. I was hoping to be able to make