Skip to main content

Table 2 Aikake Information Criterion (AIC) values from fitting models of diversification rate using the fitdAICrc function of LASER [57, 58]

From: Anatomy of a Neotropical insect radiation

Clade Lineages Best fitting model Max % missing (1-tailed)
All clades all birth-death n/a
Clade A all pure birth n/a
Clade B all DDL 10%
Clade C all pure birth n/a
All clades >  1 Ma pure birth n/a
Clade A >  1 Ma DDX ns
Clade B >  1 Ma DDX ns
Clade C >  1 Ma pure birth n/a
  1. Top four rows, “all” lineages = lineages differing by ≥ 3 observed mutations in mtCOI. Bottom four rows: results of analyses of tree truncated at 1 Ma. Constant rate models with (birth-death) and without (pure birth) extinction were compared to logarithmic (DDL) and exponential (DDX) density dependent models in which diversification rate declines as a function of the number of lineages. The “best fitting model” column identifies models with the lowest AIC value. The ΔAIC column indicates the difference in AIC values between the best fitting density dependent model and the best fitting constant rate model, such that positive values indicate a better fit to density dependent models. The maximum percent missing column shows the highest percentage of missing taxa (in 10% increments) for simulations in which critical values for ΔAIC values were lower than the observed ΔAIC value shown. NS indicates that the ΔAIC value was not significant even assuming complete taxon sampling. N/A indicates “not applicable” because the test was a one-tailed test