Hummingbirds vs Swifts

Single regime multivariate OU model (swifts and hummingbirds shared the same evolutionary regime) vs 2 regime multivariate OU model (swifts and hummingbirds have different evolutionary regimes). These models test if the different clades show distinct modes of evolution for the clade. Here the test is multivariate. Meaning that we are taking into account the covariance between the traits:

## make.simmap is sampling character histories conditioned on the transition matrix
## 
## Q =
##              humm        swift
## humm  0.000000000  0.000000000
## swift 0.001694351 -0.001694351
## (estimated using likelihood);
## and (mean) root node prior probabilities
## pi =
##  humm swift 
##   0.5   0.5

Likelihood-ratio test:

LRT(mod.2.regm, mod.1.regm)

## -- Log-likelihood Ratio Test -- 
## Model OUM  symmetric positive  versus  OU1  symmetric positive 
## Number of degrees of freedom : 3 
## LRT statistic: 8.390183  p-value: 0.03859989 * 
## --- 
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

#print best model
mod.2.regm

## 
## -- Summary results -- 
## LogLikelihood:    -694.4163 
## AIC:      1424.833 
## AICc:     1425.62 
## 18 parameters 
## 
## Estimated theta values 
## ______________________ 
##       elm.types sq.complx acous.spc
## humm  0.9791316 0.4774590  3.889852
## swift 1.0244504 0.4664176  4.396426
## 
## ML alpha values 
## ______________________ 
##           elm.types sq.complx acous.spc
## elm.types  50.29482  10.92440 -14.71905
## sq.complx  10.92440  81.37813 -16.57136
## acous.spc -14.71905 -16.57136  14.07678
## 
## ML sigma values 
## ______________________ 
##           elm.types sq.complx acous.spc
## elm.types  43.68995 16.418326 26.506826
## sq.complx  16.41833 13.532477  3.035063
## acous.spc  26.50683  3.035063 22.597034

Different regimes for each clade

Likelihood-ratio test:

LRT(mod.sev.regm, mod.1.regm)

## -- Log-likelihood Ratio Test -- 
## Model OUM  symmetric positive  versus  OU1  symmetric positive 
## Number of degrees of freedom : 24 
## LRT statistic: 68.95751  p-value: 3.14161e-06 *** 
## --- 
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

#print best model
mod.sev.regm

## 
## -- Summary results -- 
## LogLikelihood:    -662.1531 
## AIC:      1402.306 
## AICc:     1405.998 
## 39 parameters 
## 
## Estimated theta values 
## ______________________ 
##            elm.types  sq.complx acous.spc
## Bees       1.3796173 0.57955751  4.379859
## Brilliants 0.4448918 0.23554891  2.999917
## Coquettes  1.3656833 0.57824788  4.482645
## Emeralds   0.9644421 0.50703964  4.022721
## Hermits    1.0739820 0.59758598  4.017954
## Mangoes    0.7929528 0.44710163  3.428027
## Mtn. Gems  0.9529635 0.32967498  3.601510
## Swifts     1.0238807 0.46663848  4.395562
## Topazes    0.2890372 0.09593729  2.725116
## 
## ML alpha values 
## ______________________ 
##           elm.types sq.complx acous.spc
## elm.types  53.97556   4.48005 -13.78145
## sq.complx   4.48005  79.26165 -12.93177
## acous.spc -13.78145 -12.93177  13.30901
## 
## ML sigma values 
## ______________________ 
##           elm.types sq.complx acous.spc
## elm.types  42.69108 13.747617 26.891403
## sq.complx  13.74762 11.449907  2.184997
## acous.spc  26.89140  2.184997 22.908470