Table 1. Simulation check of the Bayesian (B) and classical (C) models.
M | p | Median | Mean | RMSE | ||||
---|---|---|---|---|---|---|---|---|
B | C | B | C | B | C | |||
a) | 0.5 | 0.15 | 0.499 | 0.500 | 0.501 | 0.501 | 0.028 | 0.030 |
0.2 | 0.15 | 0.501 | 0.500 | 0.501 | 0.500 | 0.022 | 0.024 | |
0.5 | 0.30 | 0.500 | 0.501 | 0.500 | 0.501 | 0.023 | 0.023 | |
0.2 | 0.30 | 0.499 | 0.499 | 0.500 | 0.500 | 0.018 | 0.018 | |
b) | 0.5 | 0.15 | 0.502 | 0.501 | 0.502 | 0.501 | 0.026 | 0.030 |
0.2 | 0.15 | 0.501 | 0.500 | 0.501 | 0.501 | 0.021 | 0.024 | |
0.5 | 0.30 | 0.499 | 0.501 | 0.500 | 0.500 | 0.021 | 0.023 | |
0.2 | 0.30 | 0.499 | 0.500 | 0.500 | 0.500 | 0.018 | 0.018 |
The data were simulated based on a) a covariate and b) a random effect, for occupancy. The true median and mean occupancy estimates were both 0.5 for all scenarios. M represents the proportions of sites missed per year and p the detection probability. RMSE denotes the root-mean-square-error.