Table 3. Unequal proportions heterogeneous scenario; a heterogeneous population with three clusters.

λ₁ = 0.475, λ₂ = 0.475, λ₃ = 0.05, μ₁ = 1, μ₂ = 2, μ₃ = 3 and σ₁ = σ₂ = σ₃ = 0.25 (high separation). Percentage of data sets in which the true number of clusters was found, with the mode of the estimated number of classes in parentheses.

α	RJMCMC	${\text{R\&M}}_0^{\text{NI}}$	${\text{R\&M}}_{0.01}^{\text{NI}}$	${\text{R\&M}}_{0.02}^{\text{NI}}$	${\text{R\&M}}_{0.05}^{\text{NI}}$	DIC3	DIC4
0.00001	—	100%(3)	100%(3)	100%(3)	48%(2)	0%(5)	14%(4)
0.001	6%(7)	98%(3)	98%(3)	98%(3)	48%(2)	2%(5)	14%(4)
0.01	36%(4)	98%(3)	98%(3)	98%(3)	50%(3)	40%(3)	24%(4)
0.05	70%(3)	100%(3)	100%(3)	100%(3)	50%(3)	78%(3)	16%(4)
0.1	98%(3)	100%(3)	100%(3)	100%(3)	50%(3)	98%(3)	20%(4)
0.3	100%(3)	100%(3)	100%(3)	100%(3)	54%(3)	98%(3)	28%(4)
0.5	100%(3)	100%(3)	100%(3)	100%(3)	56%(3)	100%(3)	44%(3)
0.9	100%(3)	20%(4)	88%(3)	96%(3)	80%(3)	100%(3)	92%(3)

The success rate of BIC using a frequentist approach was 98%(3).