[本週專題]WinBUGS-Undefined Real Result?

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

Dear lister
I am having an issue with WINBUGS when some of my priors are uniform.  When A[1,1] and A[1,2] have a uniform(0,1) prior I get the TRAP, but if I switch them to normal(0,1) it is fine. If I try to truncate the normal though, I get the same error. Any suggestion would be greatly appreciated.
Thanks

1. Code:
   
2. ## Set up A matrix ##
   
3. A[1,1] ~ dunif(0,1) ## THESE TWO ENTRIES CAUSE THE PROBLEM ##
   
4. A[1,2] ~ dunif(0,1)
   
5. A[2,1] <- 1
   
6. A[2,2] <- 0
   
7. A[3,1] <- 0
   
8. A[3,2] <- 1
   
9. 
   
10. nu.inv1 ~ dunif(0,1)
    
11. nu.inv2 ~ dunif(0,1)
    
12. sigma.beta ~ dunif(0,1000)
    
13. nu1 <- 1/nu.inv1
    
14. nu2 <- 1/nu.inv2
    
15. a1 <- nu1/2; b1 <- (nu1/2)*pow(sigma.beta,2)
    
16. a2 <- nu2/2; b2 <- (nu2/2)*pow(sigma.beta,2)
    
17. tau.beta1 ~ dgamma(a1,b1)
    
18. tau.beta2 ~ dgamma(a2,b2)
    
19. 
    
20. for (j in 1 : J){
    
21. 
    
22. ## Setup priors for betas ##
    
23. B[j,1] ~ dnorm(0,tau.beta1)
    
24. B[j,2] ~ dnorm(0,tau.beta2)
    
25. 
    
26. mu[j,3] <- B[j,2]
    
27. mu[j,2] <- B[j,1]
    
28. ## uniform priors cause issues here ##
    
29. mu[j,1] <- A[1,1]*B[j,1] + A[1,2]*B[j,2]
    
30. 
    
31. X[j,1] <- Y1[j]
    
32. X[j,2] <- Y2[j]
    
33. X[j,3] <- Y3[j]
    
34. 
    
35. X[j,1:3] ~ dmnorm(mu[j,1:3],tau[,])
    
36. }

复制代码

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：Undefined winbugs Defined DEFINE WINBUG having matrix normal error