Here is simple MVN  model for InterStim prediction problem:


		model
		{
			for (i in 1:N){
			   X[i, 1:4] ~ dmnorm(theta[1:4], tau[1:4, 1:4])
			}
			theta[1:4] ~ dmnorm(mu[1:4], Omega[1:4, 1:4])
			tau[1:4, 1:4] ~ dwish(R[1:4, 1:4], 4)
			alpha <- (theta[1] - theta[3])/theta[1]
			beta <- (theta[1] - theta[4])/theta[1]
		}
	
		
	Here is Data:
	
	list(N = 49, X = structure(
			.Data = c(60, 0.7, 0, 16,
8, 0, 0, 0,
2, 0, 1, 4,
8, 0, 0.3, 0,
7, 0, 0, 0,
3, 3, 2, NA,
9, 0.3, 0.3, 0,
9, 0.7, 12, NA,
3, 0, 0.7, 0,
3, NA, NA, NA,
22, 6, 1, 0.3,
2, 2, 0, 0,
40, NA, NA, NA,
2, 0.3, NA, NA,
50, 0, 5, NA,
7, 0, 1, NA,
4, NA, NA, NA,
19, 1, NA, NA,
4, 1, NA, NA,
10, 0, 0, 0,
10, 0, NA, NA,
12, NA, NA, 7,
3, NA, 0, 0.3,
79, 5, 6, 8,
23, 4, 8, 8,
11, 3, 3, NA,
6, 2, 2, NA,
18, 2, NA, NA,
56, NA, NA, NA,
10, 1, 0, 0,
13, 0.3, 7, NA,
6, NA, NA, NA,
8, 4, 0.7, NA,
3, 0.7, 2, NA,
7, NA, NA, NA,
5, NA, 0.3, NA,
6, 0.3, 2, NA,
22, NA, NA, NA,
8, 0.3, 1, NA,
27, NA, NA, NA,
9, 0.7, 0.7, NA,
54, 1, 2, NA,
9, 0.3, 2, NA,
22, 8, 5, 4,
7, 0, 0, 0.7,
4, 0, 0, 0,
6, 0, 0, NA,
5, 0, 3, NA,
16, NA, NA, NA),
			.Dim = c(49, 4)),
			mu = c(0, 0, 0, 0),
			R = structure(.Data = c(0.1, 0, 0, 0,
													0, 0.1, 0, 0,
													0, 0, 0.1, 0,
													0, 0, 0, 0.1), .Dim = c(4, 4)),
			Omega = structure(.Data = c(1.0E-6, 0, 0, 0, 
													0, 1.0E-6, 0, 0, 
													0, 0, 1.0E-6, 0, 
													0, 0, 0, 1.0E-6), .Dim = c(4, 4)))