Skip to main content
. 2007 Jan 30;104(6):1760–1765. doi: 10.1073/pnas.0607208104

Table 1.

Mean number of data-generation steps per final particle for each population, based on 1,000 particles, under algorithms ABC-PRC and ABC-REJ

t ϵt ABC-PRC ABC-REJ
Prior Posterior
1 2.000 4.907
2 0.500 4.899
3 0.025 66.089 400.806 21.338
Total 75.895 400.806 21.338

Final two columns use U(−10, 10) prior and known posterior mixture of normals as sampling distributions.