跪求大神帮看一下WinBUGS中的初值问题

我在贝叶斯推断的突变点研究，为什么赋初值时出错，显示this chain contains uninitialized variables，代码如下（我第一次用这个软件，还不太会）：
model
         {
                for (i in 3:N)
                        {
                        y ~ dnorm (mu, tao[J])
                        mu <- alpha[J]+beta[J]*i+gamma*x1
                        J<-1+step(i-x.change1)+step(i-x.change2)
                        x1<-y[i-1]
                        }
                for(i in 1:N-2)
                        {
                        punif<-1/(N-2)
                        }
                for(j in 1:3)
                {
                alpha[j] ~ dnorm(0,0.001)
                beta[j] ~ dnorm(0,0.001)
                tao[j] ~ dgamma(1.001,0.001)
                }
                k[1]~dcat(punif[])
                k[2]~dcat(punif[])
                x.change1<-ranked(k[],1)+1
                x.change2<-ranked(k[],1)+1
                gamma ~ dnorm(0,0.001)
        }
       
list(N=56,y=c(3209.91,3075.50,2889.01,3008.79,3206.36,3220.19,3163.76,2987.03,3056.84,3053.39,2661.87,2541.40,2493.39,2636.22,2645.26,2942.52,2973.67,2845.88,2766.58,2872.01,2941.89,2995.94,2827.07,2708.68,2773.56,2592.50,2460.17,2401.14,2456.55,2248.84,2250.15,2365.99,2383.88,2356.65,2388.54,2282.52,2168.81,2114.29,2070.15,2100.09,2042.74,2124.66,2311.04,2380.36,2296.43,2211.80,2264.51,2119.15,2013.65,2073.82,2185.35,2186.71,2161.97,2163.38,2036.28,2091.49))

list(alpha=c(0,0,0),beta=c(0,0,0),tao(0,0,0)=(0.01,0.01,0.01),gamma=0,x.change1=2,x.change2=2,
k=c(2,2))

mu <- alpha[J]+beta[J]*i+gamma*x1
--------------------------------
j未定义