1 2004 10 15 SAS 2 MCMC 3 4 ∫ = θ θ θ θ θ θ d p x p p x p x p ) ( ) | ( ) ( ) | ( ) | ( ) | ( θ x p θ ) (θ p ) | ( x p θ Bayesian inference. Encyclopedia of Statistical Science,Vol.1,pp.197-204. 5 MCMC MCMC Metropolis-Hastings 6 Metropolis-Hastings ) (⋅ π ) 0 ( x N j ,.., 2 , 1 = ) , ( ) 1 ( ⋅ − j x q y ) 1 , 0 ( U ) , ( ) 1 ( y x u j− ≤α y x j = ) ( ⎭ ⎬ ⎫ ⎩ ⎨ ⎧ = − − − − ) , ( ) ( ) , ( ) ( , 1 m