write.jags.model: Write and remove model file

In datacloning/dclone: Data Cloning and MCMC Tools for Maximum Likelihood Methods

Write and remove model file

Description

Writes or removes a BUGS model file to or from the hard drive.

Usage

write.jags.model(model, filename = "model.txt", digits = 5,
    dir = tempdir(), overwrite = getOption("dcoptions")$overwrite)
clean.jags.model(filename = "model.txt")
custommodel(model, exclude = NULL, digits = 5)

Arguments

model	JAGS model to write onto the hard drive (see Example). For write.jags.model, it can be name of a file or a function, or it can be an 'custommodel' object returned by custommodel. custommodel can take its model argument as function. If model is not function, its is coerced as character.
digits	Number of significant digits used in the output.
filename	Character, the name of the file to write/remove. It can be a link{connection}.
dir	Optional argument for directory where to write the file. The default is to use a temporary directory and use file.path(dir, filename). When NULL, it uses the current working directory (getwd()).
overwrite	Logical, if TRUE the filename will be forced and existing file with same name will be overwritten.
exclude	Numeric, lines of the model to exclude (see Details).

Details

write.jags.model is built upon the function write.model of the R2WinBUGS package.

clean.jags.model is built upon the function file.remove, and intended to be used internally to clean up the JAGS model file after estimating sessions, ideally via the on.exit function. It requires the full path as returned by write.jags.model.

The function custommodel can be used to exclude some lines of the model. This is handy when there are variations of the same model. write.jags.model accepts results returned by custommodel. This is also the preferred way of including BUGS models into R packages, because the function form often includes undefined functions.

Use the %_% operator if the model is a function and the model contains truncation (I() in WinBUGS, T() in JAGS). See explanation on help page of write.model.

Value

write.jags.model invisibly returns the name of the file that was written eventually (possibly including random string). The return value includes the full path.

clean.jags.model invisibly returns the result of file.remove (logical).

custommodel returns an object of class 'custommodel', which is a character vector.

Author(s)

Peter Solymos

See Also

write.model, file.remove

Examples

## Not run: 
## simple regression example from the JAGS manual
jfun <- function() {
    for (i in 1:N) {
        Y[i] ~ dnorm(mu[i], tau)
        mu[i] <- alpha + beta * (x[i] - x.bar)
    }
    x.bar <- mean(x)
    alpha ~ dnorm(0.0, 1.0E-4)
    beta ~ dnorm(0.0, 1.0E-4)
    sigma <- 1.0/sqrt(tau)
    tau ~ dgamma(1.0E-3, 1.0E-3)
}
## data generation
set.seed(1234)
N <- 100
alpha <- 1
beta <- -1
sigma <- 0.5
x <- runif(N)
linpred <- crossprod(t(model.matrix(~x)), c(alpha, beta))
Y <- rnorm(N, mean = linpred, sd = sigma)
## list of data for the model
jdata <- list(N = N, Y = Y, x = x)
## what to monitor
jpara <- c("alpha", "beta", "sigma")
## write model onto hard drive
jmodnam <- write.jags.model(jfun)
## fit the model
regmod <- jags.fit(jdata, jpara, jmodnam, n.chains = 3)
## cleanup
clean.jags.model(jmodnam)
## model summary
summary(regmod)

## End(Not run)
## let's customize this model
jfun2 <- structure(
    c(" model { ",
    "     for (i in 1:n) { ",
    "         Y[i] ~ dpois(lambda[i]) ",
    "         Y[i] <- alpha[i] + inprod(X[i,], beta[1,]) ",
    "         log(lambda[i]) <- alpha[i] + inprod(X[i,], beta[1,]) ",
    "         alpha[i] ~ dnorm(0, 1/sigma^2) ",
    "     } ",
    "     for (j in 1:np) { ",
    "         beta[1,j] ~ dnorm(0, 0.001) ",
    "     } ",
    "     sigma ~ dlnorm(0, 0.001) ",
    " } "),
    class = "custommodel")
custommodel(jfun2)
## GLMM
custommodel(jfun2, 4)
## LM
custommodel(jfun2, c(3,5))
## deparse when print
print(custommodel(jfun2), deparse=TRUE)

datacloning/dclone documentation built on Sept. 29, 2024, 3:21 p.m.