CIRCMLE |
OPTM - Beta testing |
The R package 'CircMLE' is a series of wrapper functions to implement the 10 maximum likelihood models of animal orientation described by Schnute and Groot (1992) <doi:10.1016/S0003-3472(05)80068-5>. The functions also include the ability to use different optimizer methods, calculate various model selection metrics (i.e., AIC, AICc, BIC), perform likelihood ratio tests and plot results.
|
The R package 'OptM' is a tool I am developing to determine the optimal number of migration edges on a population tree using the treemix software. Currently, treemix recommends to stop adding migration edges when 99.8% of the observed variation is explained by the model. Using OptM, one can use either an ad hoc statistic based on the second-order rate of change in likelihood (what I call "Δm") or a series of linear modeling approaches to determine the optimal cutoff using a more quantitative approach.
OptM is currently available on CRAN in a beta-testing phase. Please check it out and let me know how it can improve! |