Table 2.

Temporal lobe epilepsy dataset: (a) Posterior mean of β; (b) 95% credible interval (CI) for β; and (c) posterior probability of odds ratio >1, e.g., $\mathbb{P}[e^{{\beta }_j}>1|X,Y]=\mathbb{P}[{\beta }_j>0|X,Y]$, shown for proposed approach (f = 0.1), multi-step logistic approach, and multi-step sparse clustering approach.

	Proposed Method			Multistep logistic regression			Multistep sparse clustering
	(a)	(b)	(c)	(a)	(b)	(c)	(a)	(b)	(c)
β0	−0.377	(−1.588, 3.587)	0.43	−0.225	(−0.8, 1.665)	0.41	−0.169	(−0.746, 1.718)	0.43
β1	0.368	(0.275, 0.685)	0.99	0.250	(0.183, 0.48)	0.99	0.262	(0.196, 0.486)	0.99
β2	−3.726	(−4.639, −0.76)	0.01	−1.25	(−1.742, 0.347)	0.06	−1.247	(−1.737, 0.358)	0.06
β3	−0.082	(−0.117, 0.032)	0.09	−0.069	(−0.094, 0.011)	0.05	−0.079	(−0.107, 0.014)	0.05
β4	1.649	(0.832, 4.364)	0.90	0.939	(0.453, 2.552)	0.88	0.41	(−0.113, 2.140)	0.68

Here β₁ ≡ Epilepsy duration, β₂ ≡ History of GTC, β₃ ≡ Age at surgery, β₄ ≡ Cluster 1 (v. Cluster 2).

Odds are with respect to seizure freedom.