[本週專題]R2WinBUGS: 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 R users,

Attached please review view my log file. I am using R2WinBUGS to call WinBUGS from R. After loading data, model, and initial values I called the command as follows and got some inf parameter values. I do not understand why this happens as the program was working fine for 5*10^5 iterations. I tried to change the seed values but of no use.

1. res <- bugs(data = dfile, inits = list(ifile), parameters.to.save = c("beta"), model.file = mfile, working.directory = tdir, n.thin = nthin, n.chains = 1, n.iter = niterations*nthin, n.burnin = 0, DIC = F, debug = T, save.history=F)

复制代码

1. display(log)
   
2. check(C:/Documents and Settings/sbadamakis/Desktop/model_4/model_4.txt)
   
3. model is syntactically correct
   
4. data(C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/data.txt)
   
5. data loaded
   
6. compile(1)
   
7. model compiled
   
8. inits(1,C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/inits1.txt)
   
9. model is initialized
   
10. gen.inits()
    
11. command #Bugs:gen.inits cannot be executed (is greyed out)
    
12. thin.updater(1)
    
13. update(0)
    
14. set(beta)
    
15. update(20)
    
16. coda(*,C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/coda)
    
17. stats(*)
    
18. Node statistics
    
19. node mean sd MC error 2.5% median 97.5% start sample
    
20. beta[1] 1.588 0.009812 0.003936 1.579 1.584 1.607 1 20
    
21. beta[2] -0.04272 0.00263 0.001137 -0.04592 -0.0433 -0.03688 1 20
    
22. beta[3] -0.07878 8.531E-4 2.829E-4 -0.07962 -0.07833 -0.07699 1 20
    
23. beta[4] -0.3035 0.01016 0.004509 -0.3194 -0.3028 -0.2841 1 20
    
24. beta[5] 1.241 0.00854 0.002739 1.228 1.242 1.258 1 20
    
25. beta[6] -4.155 0.01242 0.005405 -4.174 -4.152 -4.132 1 20
    
26. beta[7] 0.08322 0.001485 6.379E-4 0.08201 0.0824 0.08608 1 20
    
27. beta[8] -3.001 0.009335 0.003318 -3.013 -3.004 -2.981 1 20
    
28. beta[9] 0.07953 7.547E-4 3.198E-4 0.07777 0.07932 0.08035 1 20
    
29. save(C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/log.odc)
    
30. save(C:/Documents and Settings/sbadamakis/Desktop/model_4/output1/log.txt)
    
31. So far so good. After the end of the last line of the log file, WinBUGS should close and the bugs function in R should return the Markov chain values of the "beta" parameters. The problem is that it doesn't and WinBUGS stays infinitely open. This is a big problem as I have written a code in R and I want to call the bugs function several times. Do you know if there's an option in the bugs function in R to FORCE WinBUGS to close?
    
32. Furthermore, some other times (the same code!!!) produces the following TRAP message
    
33. undefined real result
    
34. GraphStack.Node.Value [00000C9AH]
    
35. .i INTEGER 6
    
36. .j INTEGER 2
    
37. .k INTEGER 1
    
38. .node GraphStack.Node [0138C5D0H]
    
39. .op INTEGER 6
    
40. .p GraphNodes.Node [01544D30H]
    
41. .top INTEGER 0
    
42. .work ARRAY 10 OF REAL
    
43.    [0] REAL inf   [1] REAL inf   [2] REAL inf   [3] REAL 1.30212336033939   [4] REAL 4.534462964264759   [5] REAL 1.769649991139459E+167   [6] REAL 1.792592453177908E-307   [7] REAL 9.94060079432588E-321   [8] REAL 5.474495960374562E+67   [9] REAL 5.257527858867595E-308
    
44. MonitorsSamples.StdMonitor.Update [0000023DH]
    
45. .blocks POINTER NIL
    
46. .i INTEGER 1655837286
    
47. .index INTEGER 7
    
48. .monitor MonitorsSamples.StdMonitor [02155B50H]
    
49. .numBlocks INTEGER 0
    
50. .value REAL 1.0
    
51. 
    
52. SamplesMonitors.Monitor.Update [0000091BH]
    
53. .chain INTEGER 0
    
54. .i INTEGER 4480
    
55. .monitor SamplesMonitors.Monitor [01C0A4E0H]
    
56. .size INTEGER 6780
    
57. 
    
58. SamplesIndex.Monitor.Update [0000030CH]
    
59. .chain INTEGER 0
    
60. .cursor SamplesIndex.List [01AB2E30H]
    
61. .monitor SamplesIndex.Monitor [01C59510H]
    
62. .sampleMonitor SamplesMonitors.Monitor [01C0A4E0H]
    
63. UpdaterActions.UpdateMonitors [0000011CH]
    
64. .chain INTEGER 0
    
65. .monitor UpdaterActions.Monitor [01C59510H]
    
66. BugsCmds.Action.Step [00000BCAH]
    
67. .a BugsCmds.Action [011D40C0H]
    
68. .chain INTEGER 0
    
69. .i INTEGER 8
    
70. .j INTEGER 1000
    
71. .numChains INTEGER 1
    
72. 
    
73. BugsCmds.Action.Do [00000C58H]
    
74. .a BugsCmds.Action [011D40C0H]
    
75. .elapsedTime LONGINT 62EC21F162EB0384H
    
76. .profileList ARRAY 2024 OF CHAR "" ...
    
77. .res INTEGER 1635734111
    
78. .s ARRAY 120 OF CHAR 3X ...
    
79. 
    
80. Services.Exec [00000136H]
    
81. .a Services.Action [011D40C0H]
    
82. .t POINTER [62EB0384H]
    
83. 
    
84. Services.IterateOverActions [000002F4H]
    
85. .p Services.Action [011D40C0H]
    
86. .t POINTER NIL
    
87. .time LONGINT 300843187
    
88. 
    
89. Services.StdHook.Step [0000034DH]
    
90. .h Services.StdHook [0101E380H]
    
91. HostWindows.Idle [00004A86H]
    
92. .focus BOOLEAN FALSE
    
93. .tick Controllers.TickMsg Fields
    
94. .w HostWindows.Window NIL
    
95. HostMenus.TimerTick [00003422H]
    
96. .lParam INTEGER 0
    
97. .ops Controllers.PollOpsMsg Fields
    
98. .wParam INTEGER 1
    
99. .wnd INTEGER 1770006
    
100. Kernel.Try [00003A61H]
     
101. .a INTEGER 1770006
     
102. .b INTEGER 1
     
103. .c INTEGER 0
     
104. .h PROCEDURE HostMenus.TimerTick
     
105. HostMenus.ApplWinHandler [00003841H]
     
106. .Proc PROCEDURE NIL
     
107. .hit BOOLEAN FALSE
     
108. .lParam INTEGER 0
     
109. .message INTEGER 275
     
110. .res INTEGER 4243644
     
111. .s ARRAY 256 OF SHORTCHAR "ˉB", 1X ...
     
112. .w INTEGER 2147348480
     
113. .wParam INTEGER 1
     
114. .wnd INTEGER 1770006
     
115. <system>   (pc=7E418733H,  fp=0022FB84H)<system>   (pc=7E418815H,  fp=0022FBECH)<system>   (pc=7E4189CCH,  fp=0022FC4CH)<system>   (pc=7E4196C6H,  fp=0022FC5CH) HostMenus.Loop [00003BDEH]
     
116. .done BOOLEAN FALSE
     
117. .f SET {0..5}
     
118. .n INTEGER 0
     
119. .res INTEGER 0
     
120. .w HostWindows.Window NIL
     
121. Kernel.Start [00002B8CH]
     
122. .code PROCEDURE HostMenus.Loop
     
123. Is there anyone who can explain what does this mean and how to understand trap messages? For me, it seems that the problem lies here:
     
124. .work ARRAY 10 OF REAL
     
125.    [0] REAL inf   [1] REAL inf   [2] REAL inf   [3] REAL 1.30212336033939   [4] REAL 4.534462964264759   [5] REAL 1.769649991139459E+167   [6] REAL 1.792592453177908E-307   [7] REAL 9.94060079432588E-321   [8] REAL 5.474495960374562E+67   [9] REAL 5.257527858867595E-308

复制代码

Notes:

• I am running multilevel models for small area estimation and the above problems seem to appear only with the spatial random effects model. The spatial and non-spatial random effects model seems to behave properly without any problems.
• The program was working fine for more than 500000 iterations withoud trap messages.
  

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 R2WinBUGS Defined winbugs WINBUG understand command loading initial change

Without having your model file etc,  I can guess that you have some division by almost zero from sampling from some distribution that may return some extremely small values close to zero.

Do you know if there's an option in the bugs function in R to FORCE WinBUGS to close?
You can close WinBUGS and everything should be fine - or just set debug = FALSE rather than debug = TRUE in your bugs() call.

Most of this type problems come from the prior. Check whether your prior range is larger than your data.

scure 发表于 2014-6-28 10:27
Most of this type problems come from the prior. Check whether your prior range is larger than your d ...

The BUGS mannual also provides other explanations for this problem. You can check it yourself.
http://mathstat.helsinki.fi/open ... g.html#TrapMessages