Table 2. Model comparison and parameter values for the best fit of each model.
| Model | LLH | AIC | Δ i | wi | Kp | Ks | dd | dp | D | σ | Dm | H | Sd | Sp |
| Complete (i,j,k,l) | −712.1 | 1442.2 | 1.78 | 0.21 | 55.3 | 43.6 | 0.36 | 0.30 | 38.9 | 31.0 | 57.4 | 14.7 | 0.50 | 0.95 |
| Sylvatic only (ii) | −733.6 | 1483.1 | 42.7 | ≈10−10 | 0 | 52.4 | 0.02 | 0.06 | 14.6 | 247.9 | 259.6 | 35.9 | 0.68 | 0.96 |
| Peridomestic only (iii) | −722.7 | 1461.4 | 21.1 | ≈10−5 | 51.7 | 0 | 0.42 | 0.22 | 10.7 | 37.1 | 44.7 | 72.0 | 0.65 | 0.98 |
| Same departure probability (jj) | −712.2 | 1440.4 | 0 | 0.52 | 69.4 | 56.1 | 0.38 | 0.38 | 49.3 | 17.7 | 55.5 | 10.5 | 0.37 | 0.97 |
| Null distance mode (kk) | −713.0 | 1442.1 | 1.68 | 0.22 | 59.9 | 47.8 | 0.91 | 0.42 | 0 | 44.6 | 47.2 | 16.4 | 0.44 | 0.97 |
| Random dispersal (ll) | −714.7 | 1445.3 | 4.95 | 0.04 | 198.6 | 241.7 | 0.15 | 0.14 | 60.5 | 9.5 | 62.2 | 1 | 0.74 | 0.81 |
Hypotheses and parameters are as described in Table 1. In addition, Dm, the mean dispersal distance (in meters), has been calculated from the estimates D and σ. Along with the name of each model, the first column includes a summary of the underlying hypotheses. The ‘Complete model’ corresponds to hypotheses i,j,k,l, which is noted (i,j,k,l). Each of the other models is defined by changing one of those hypotheses, and only this changed hypothesis is reported. The best model and corresponding optimal parameter values appear in bold. Boxed cells indicate that the parameter values have been constrained according to the hypotheses being considered.