Table 2.

Performance of old and new likelihood calculations for several combinations of N, μ, and s.

			Default Method		New Method
Population size (N)	θ	Ns	MPE	Fraction containing true value	MPE	Fraction containing true value
500	0.01	0	470 (93)	0.98	537 (185)	0.87
500	0.01	10	317 (65)	0.14	475 (192)	0.96
500	0.05	20	187 (38)	0.0	387 (116)	0.94
1000	0.01	0	980 (233)	0.95	1088 (343)	0.87
1000	0.01	10	647 (137)	0.2	1019 (356)	0.97
1000	0.025	20	486 (104)	0.0	922 (280)	0.95
1000	0.05	25	365 (68)	0.0	804 (225)	0.9
2500	0.025	0	2578 (515)	0.95	2880 (830)	0.87
2500	0.025	10	1270 (283)	0.0	2443 (884)	0.96
2500	0.025	25	1315 (243)	0.0	1902 (701)	1
5000	0.05	10	2635 (557)	0.0	5055 (1711)	0.97
5000	0.025	25	2467 (501)	0.0	4221 (1481)	0.98

MPE, Maximum a posteriori estimate. Fraction containing true value indicates the fraction of data sets in which the 95% CPD contained the true population size.