Skip to main content
. 2016 Feb 18;6:21323. doi: 10.1038/srep21323

Table 2. AICc (Akaike information criterion adjusted for sample size) scores of all candidate generalized additive models (GAMs).

 Candidate models Dependent variable
Abundance of infected Infection prevalence Seroconversion rate
Within-cycle dynamics
 cycle ID + s(cycle month by cycle ID) 626.4 543.3 169.1
 cycle ID + s(cycle month) 646.9 533.2 186.0
 s(cycle month) 762.9 530.1 184.3
 cycle ID 643.9 953.7 193.9
 intercept only 761.0 951.2 196.5
Within-year dynamics
 cycle phase + s(month by cycle phase) 578.1 507.4 134.6
 cycle phase + s(month) 600.1 490.5 134.1
 s(month) 610.3 497.2 136.5
 cycle phase 1374.8 549.0 132.1
 intercept only 1381.2 550.9 135.5
Variation in relation to breeding season (dichotomous)
 cycle phase * breeding season     130.4
 cycle phase + breeding season     127.9
 breeding season     130.4
 cycle phase     130.8
 intercept only     134.1
Variation in relation to cohort age and birth year cycle phase
 cycle phase + s(cohort age by cycle phase)   183.3 156.1
 cycle phase + s(cohort age)   182.2 158.3
 s(cohort age)   205.3 160.9
 cycle phase   519.4 158.7
 intercept only   642.2 161.9

s(…) denote GAM smooth terms.

The AICc scores of the best-supported models are written in bold.