Skip to main content

Table 1 The ten best models according to Akaike’s Information Criteria (AIC)

From: Coevolution of group-living and aposematism in caterpillars: warning colouration may facilitate the evolution from group-living to solitary habits

 

K

logLik

AIC

\(\Delta\)AIC

LikRatio

AkaikeWeight

n1

7

\(-\)362.705

739.410

0.000

1.000

0.201

eq24

7

\(-\)362.754

739.508

0.099

0.952

0.192

eq13

7

\(-\)362.765

739.530

0.120

0.942

0.190

eq18

6

\(-\)364.216

740.431

1.022

0.600

0.121

General

8

\(-\)362.705

741.410

2.000

0.368

0.074

eq38

7

\(-\)363.804

741.609

2.199

0.333

0.067

eq16

7

\(-\)364.034

742.068

2.659

0.265

0.053

n8

7

\(-\)364.216

742.431

3.022

0.221

0.044

eq47

7

\(-\)364.674

743.347

3.938

0.140

0.028

eq57

7

\(-\)365.874

745.748

6.338

0.042

0.009

  1. \(\Delta\)AIC= difference in AIC between each model and the best model (AIC-AIC\(_{min}\)); LikRatio = likelihood ratio between each model and the best model (exp((AIC\(_{min}\)-AIC)/2)), sometimes called the ’evidence ratio’ and gives the strength of evidence for each model as a proportion of the best model; AkaikeWeight = model probabilities (probablility of each model being the best model in the set) [31, 32]