From: Using multi-step proposal distribution for improved MCMC convergence in Bayesian network structure learning
Proposal distribution
Mean DAGs (std)
p 1=1.0,p 2=0.0
4.41 (3.3698)
p 1=0.8,p 2=0.2
8.3 (2.8373)
p 1=0.6,p 2=0.4
8.01 (2.4058)
p 1=0.4,p 2=0.6
7.24 (3.0021)
p 1=0.2,p 2=0.8
6.31 (3.212)
p 1=0.0,p 2=1.0
2.66 (2.4994)