Skip to main content
. 2018 Jan 29;6(1):coy001. doi: 10.1093/conphys/coy001

Table 5a:

Comparisona of candidate modelsb by Aikaike Information Criteria (AIC) model selection to predict the age class for 86 male brown bears. Males < 3 years old were classified as immature (N = 42) whereas males ≥3 years old were classified as adult (N = 44). Model M1 is an intercept-only (null) model

Model Hormone Time Study area Contact method Interactions K AICC ∆AICC wi Accuracy (%)
M4 test, prog, est3, cort, cort3 d (prog × est3), (cort × d) 9 64.87 0.00 0.97 88.4
M6 test, prog, est3 cort, cort3 sa (prog × est3), (cort × sa), (cort3 × sa) 10 73.59 8.72 0.01 86.1
M5 test, est3 cm (est3 × cm) 7 74.39 9.52 0.01 84.9
M3 test, prog, est3, cort, cort3 (test × est3), (prog × est3) 8 74.44 9.57 0.01 84.9
M2 test, prog, est3, cort, cort3 6 80.29 15.42 0.00 86.1
M1 1 121.22 56.35 0.00 50.0

aStatistics are number of estimable parameters in model (K), sample-size-adjusted Akaike information criterion (AICC), difference in AICC between top model and model i (∆AICC), Akaike weight for model i (wi) and accuracy (%).

bVariables are testosterone (test), progesterone (prog), estradiol (est), cortisol (cort), ordinal day (d), study area (sa) and contact method (cm).