model {
for (i in 1:23) {
y[i] ~ dnorm(mu[i], tau)
D2[i] <- equals(DIST[i], 2)
D3[i] <- equals(DIST[i], 3)
mu[i] <- beta0 + beta[1]*MAN[i]
+ beta[2]*D2[i] + beta[3]*D3[i]
}
beta0 ~ dnorm(0, 0.0001)
for (j in 1:3) {
beta[j] ~ dnorm(0, 0.0001)
}
tau <- 1/pow(sigma, 2)
sigma ~ dunif(0, 100)

dummy <- AUTO[1]
}

Inits:
list(beta0 = 0, beta = c(0, 0, 0), sigma = 1)

Data:
MAN[] AUTO[] y[] DIST[]
-15.76 1.09 3.19 1
0.98 0.62 -3.45 1
3.71 0.61 0.04 1
-5.37 -1.01 6.62 1
-10.23 -0.76 3.61 1
-8.32 1.91 2.67 1
-7.80 0.40 -2.45 1
6.77 -1.71 9.31 1
-8.81 -0.76 15.29 1
-9.56 -1.34 3.68 1
-2.06 -1.71 8.63 2
-0.76 -1.82 10.82 2
-6.30 -4.91 -0.50 2
39.40 -2.65 -11.00 2
-10.79 0.11 2.05 2
-8.16 0.52 11.80 2
-2.82 -2.54 -2.02 2
-16.19 -0.07 0.94 3
-11.00 -0.83 4.42 3
-14.60 0.98 -0.86 3
-17.96 -3.41 -0.92 3
0.76 2.97 2.61 3
-10.77 2.92 1.58 3
END

node mean sd MC error 2.5% median 97.5% start sample
beta[1] -0.2378 0.1188 0.001181 -0.4759 -0.2374 -0.006211 5001 10000
beta[2] 0.558 3.027 0.03123 -5.493 0.5386 6.595 5001 10000
beta[3] -4.03 3.13 0.03011 -10.14 -4.014 2.239 5001 10000
beta0 2.573 1.992 0.01811 -1.291 2.581 6.533 5001 10000
sigma 5.837 1.037 0.01537 4.243 5.686 8.233 5001 10000