Table 2.
Posterior parameter estimates (median and 95% confidence intervals) for the best supported scenario (scenario 2, see Figure 5). Estimates are based on 1% of simulated data sets closest to the observed values. Relative median absolute errors (RMAE) based on 500 pseudo‐observed data sets are also indicated for each parameter
| Parameter | Median | q 0.025 | q 0.975 | RMAE |
|---|---|---|---|---|
| N1 | 520,000 | 194,000 | 736,000 | .279 |
| N2 | 407,000 | 113,000 | 715,000 | .246 |
| N3 | 600,000 | 277,000 | 740,000 | .258 |
| N1anc | 303,000 | 39,800 | 690,000 | .394 |
| N2‐3 | 345,000 | 46,600 | 695,000 | .372 |
| Nx | 57,000 | 24,800 | 541,000 | .384 |
| t 1 | 42,700 | 5,540 | 165,000 | .404 |
| t 2 | 215,000 | 67,600 | 342,000 | .213 |
| μ | 7.75 × 10−6 | 4.24 × 10−6 | 2.85 × 10−5 | .357 |
N1, effective population size of group A; N2, effective population size of group B; N3, effective population size of group C; N1anc, effective population size of the ancestral group A; N2‐3, effective population size of the ancestral groups B‐C; Nx, effective population size of the most ancestral population, t 1, time (in generations = years) to the most recent divergence event; t 2, time (in generations = years) to the most ancient divergence event (see scenarios in Figure 5); μ, mean mutation rate.