Table 3.
Model | Model specification | logL | AICc | ΔAIC | w i |
---|---|---|---|---|---|
bisse.null | hidden.states=FALSE, rates=c(1,1,0,0), N=4 | −476.75 | 961.77 | 30.55 | 2.1E-07 |
bisse | hidden.states=FALSE, rates=(1,2,0,0), N=6 | −471.12 | 954.84 | 19.04 | 6.7E-05 |
null.hisse | hidden.states=TRUE, rates=c(1,1,2,2), N=5 | −462.70 | 935.8 | 4.58 | 0.0919 |
hisse.partial | hidden.states=TRUE, rates=c(1,2,0,3), N=10 | −455.82 | 931.22 | 0 | 0.9079 |
The best model fitted, based on ΔAIC and Akaike weights (wi), is denoted in bold. AICc: Akaike information criterion corrected for sample size. N = number of parameters of each model; logL = log-likelihood of the model.