OpenBUGS322到R的通信

Question

我有一个关于尝试使用R2OpenBUGS使OpenBUGS322与R一起运行的基本问题。

在运行多个试用脚本时(参见下面的一个示例)，我得到一个错误

矩阵(，n.sims，n.parameters)中出错:无效的'nrow‘值(太大或NA)

通过搜索，我发现其他人在使用OpenBUGS322时遇到了这种情况，OpenBUGS321应该可以正常工作。但是，有没有人可以建议我在哪里可以找到旧版本。我需要OpenBUGS321setup.exe，但是到处都找不到它。

(我已经修复了其他几个问题，比如dbus配置，当XQuartz提示时，OpenBUGS现在就会打开)

这里有更多的信息，看看我是否在正确的道路上。

我运行的版本是: Mac 10.5.8 Wine 1.4.1 XQuartz 2.6.3 OpenBUGS322

WinBUGS教科书中添加了WINE和OpenBUGS路径的试用代码(不是100%确定是否正确)

非常感谢你的帮助。

脚本：

y10<-rnorm(n=10, mean=600, sd=30)
y1000<-rnorm(n=1000, mean=600, sd=30)
summary(lm(y1000~1))
library(R2OpenBUGS)

#setwd("/Users/Pavel/wine/ProgramFiles/OpenBUGS322")
setwd("/Users/Pavel/Documents/R/OpenBUGS")
sink("model.txt")
 cat("
model{
population.mean~dunif(0,5000)
precision<-1/population.variance
population.variance<-population.sd* population.sd
 population.sd~dunif(0,100)
for(i in 1:nobs){
 mass[i]~dnorm(population.mean, precision)
}
}
", fill=TRUE)
 sink()

WINE="/opt/local/bin/wine"
WINEPATH="/opt/local/bin/winepath"
OpenBUGS.pgm="/Users/Pavel/wine/ProgramFiles/OpenBUGS322/OpenBUGS.exe"

#Package the data to be handed to OpenBUGS
win.data<-list(mass=y1000, nobs=length(y1000))

#Function to generate starting values
inits<-function()
list(population.mean=rnorm(1,600), population.sd=runif(1,1,30))

#Parameters to be monitored
params<-c("population.mean", "population.sd", "population.variance")

#MCMC settings
nc<-3     #Number of chains
ni<-1000  #Number of draws for each chain
nb<-1     #number of draws to discard as burn -in
nt<-1     #Thinning rate

out<-bugs(data=win.data, inits=inits, parameters.to.save=params, model.file="model.txt",    n.thin=nt, n.chains=nc, n.burnin=nb, n.iter=ni, OpenBUGS.pgm=OpenBUGS.pgm, WINE=WINE,     WINEPATH=WINEPATH,useWINE=T)

Answer 1

你能用rjags代替吗？语法几乎相同：

require(rjags)

modelstring <- "
    model {
        for(i in 1:nobs){
            mass[i] ~ dnorm(m, prec) # precision
        }
        m ~ dunif(0, 5000)
        prec <- 1/sqrt(SD) # convert to Std Deviation
        SD ~ dunif(0, 100)
    }
"
writeLines(modelstring,con="model.txt")

y1000 <- stats::rnorm(n=1000, mean=600, sd=30)

dataList <- list(
    mass = y1000,
    nobs = length(y1000)
    )

initsList <-  list(
    m = stats::rnorm(n=1, mean=600, sd=1),
    SD = stats::runif(n=1, min=1, max=30)
    )

parameters <- c("m","SD") # to be monitored.
adaptSteps <- 100         #  "tune" the samplers.
burnInSteps <- 100        #  "burn-in" the samplers.
nChains <- 3              # Number of chains to run.
numSavedSteps <-2000      # Total number of steps in chains to save.
thinSteps <- 1            # Number of steps to "thin" (1=keep every step).
nPerChain <- ceiling(( numSavedSteps * thinSteps ) / nChains) # Steps per chain

jagsModel <- rjags::jags.model(
    "model.txt", data=dataList,
    inits=initsList, n.chains=nChains,
    n.adapt=adaptSteps)

stats::update(jagsModel, n.iter=burnInSteps)

MCMC1 <- as.matrix(rjags::coda.samples(
    jagsModel, variable.names=parameters,
    n.iter=nPerChain, thin=thinSteps))

SDsample <- matrix(MCMC1[,grep("SD",colnames(MCMC1))],
                ncol=length(grep("SD",colnames(MCMC1))))

然后，您可以使用以下命令转换为方差：

SDsample^2

如果你正在寻找OpenBUGS321setup.exe，你应该能够找到它here。我注意到它还没有在葡萄酒中得到很好的测试。Linux仿真器是可能的吗？