nimbleNoBounds (version 1.0.2)

dLogHalfflat: Log transformed half-flat distribution .

Description

dLogHalfflat and rLogHalfflat provide a log-transformed half-flat distribution. Note, both dhalfflat and dLogHalfflat are improper. Thus, rLogHalfflat returns NAN, just as rhalfflat does.

Usage

dLogHalfflat(x, log = 0)
rLogHalfflat(n = 1)

Value

A value proportional to the density, or the log of that value, of x, such that x = log(y) and y ~ dhalfflat().

Arguments

x

A continuous random variable on the real line, where y=exp(x) and y ~ dhalfflat().

log

Logical flag to toggle returning the log density.

n

Number of random variables. Currently limited to 1, as is common in nimble. See ?replicate for an alternative. Note, NAN will be returned because distribution is improper.

Author

David R.J. Pleydell

Examples

oldpar <- par()
par(mfrow=n2mfrow(2))
## Plot 1
curve(dhalfflat(x), -11, 11, n=1001, col="red", lwd=3, xlab="y = exp(x)", ylab="dhalfflat(y)")
## Plot 2: back-transformed
curve(dLogHalfflat(x), -5, 1.5, n=1001, col="red", lwd=3, , xlab="x = log(y)")
abline(v=0:1, h=c(0,1,exp(1)), col="grey")
par(oldpar)