Change example in documentation?

[larryshamalama]

nimbleHMC/nimbleHMC/R/HMC_samplers.R

Lines 30 to 52 in 1e336ea

	## #' @examples
	## #' code <- nimbleCode({
	## #' b0 ~ dnorm(0, 0.001)
	## #' b1 ~ dnorm(0, 0.001)
	## #' sigma ~ dunif(0, 10000)
	## #' for(i in 1:N) {
	## #' mu[i] <- b0 + b1 * x[i]
	## #' y[i] ~ dnorm(mu[i], sd = sigma)
	## #' }
	## #' })
	## #'
	## #' N <- 10
	## #' constants <- list(N = N, x = 1:N)
	## #' data <- list(y = 1:N)
	## #' inits <- list(b0 = 1, b1 = 0.1, sigma = 1)
	## #'
	## #' Rmodel <- nimbleModel(code, constants, data, inits, buildDerivs = TRUE)
	## #'
	## #' conf <- configureMCMC(Rmodel, nodes = NULL)
	## #'
	## #' conf$addSampler(target = c('b0', 'b1', 'sigma'), type = 'HMC')
	## #'
	## #' Rmcmc <- buildMCMC(conf)

(same example as in the reference manual)

I tried this example at first and, while it runs well, I was a bit confused as to what it is estimating. While the posterior plots seem confusing at first, the posterior draws have little stochasticity, since the data does not have any randomness as well. Would it be better to use a more "illustrative" generated data? E.g.

N <- 1000
x <- rnorm(N)
y <- 1 + 2 * x + rnorm(N)

# nimble model

constants <- list(N = N, x = x)
data <- list(y = y)

# etc.

CC openjournals/joss-reviews#6745

[danielturek]

@larryshamalama Agreed, the example in the roxygen was not ideal. I've just changed the HMC example to something similar to your suggestion, the relevant code:

set.seed(0)
N <- 100
x <- rnorm(N)
y <- 1 + 0.3*x + rnorm(N)
constants <- list(N = N, x = x)
data <- list(y = y)

This change is made on the master branch of the nimbleHMC repository, and also merged into the joss branch for this review.

Thanks for the helpful suggestion. I'm going ahead and closing this issue.