Package 'JAGSmodule'

Load the internal JAGS module provided by runjags

Description

The runjags package contains a JAGS extension module that provides several additional distributions for use within JAGS (see details below). For now, this is duplicated within the JAGSmodule package for testing and demonstration purposes.

Usage

load_JAGSmodule()

unload_JAGSmodule()

Details

This module provides the following distributions for JAGS:

PARETO TYPE I: dpar1(alpha, sigma)

$p(x) = \alpha \sigma^{\alpha} x^{-\left(\alpha+1 \right)}$

$\alpha > 0, \sigma > 0, x > \sigma$

PARETO TYPE II: dpar2(alpha, sigma, mu)

$p(x) = \frac{\alpha}{\sigma} \left( \frac{\alpha + x - \mu}{\sigma}\right)^{-\left(\alpha+1\right)}$

$\alpha > 0, \sigma > 0, x > \mu$

PARETO TYPE III: dpar3(sigma, mu, gamma)

$p(x) = \frac{\frac{x-\mu}{\sigma}^{\frac{1}{\gamma}-1} \left(\frac{x-\mu}{\sigma}^{\frac{1}{\gamma}} +1\right)^{-2}}{\gamma \sigma}$

$\sigma > 0, \gamma > 0, x > \mu$

PARETO TYPE IV: dpar4(alpha, sigma, mu, gamma)

$p(x) = \frac{\alpha \frac{x-\mu}{\sigma}^{\frac{1}{\gamma}-1} \left(\frac{x-\mu}{\sigma}^{\frac{1}{\gamma}} +1\right)^{-\left(\alpha+1\right)}}{\gamma \sigma}$

$\alpha > 0, \sigma > 0, \gamma > 0, x > \mu$

LOMAX: dlomax(alpha, sigma)

$p(x) = \frac{\alpha}{\sigma} \left(1 + \frac{x}{\sigma}\right)^{-\left(\alpha+1\right)}$

$\alpha > 0, \sigma > 0, x > 0$

GENERALISED PARETO: dgenpar(sigma, mu, xi)

$p(x) = \frac{1}{\sigma} \left(1 + \xi \left(\frac{x-\mu}{\sigma}\right)\right)^{-\left(\frac{1}{\xi}+1\right)}$

For $\xi=0$:

$p(x) = \frac{1}{\sigma} e^{\frac{-\left(x-\mu\right)}{\sigma}}$

$\sigma > 0, x > \mu$

DUMOUCHEL: dmouch(sigma)

$p(x) = \frac{\sigma}{\left(x+\sigma\right)^2}$

$\sigma > 0, x > 0$

HALF CAUCHY: dhalfcauchy(sigma)

$p(x) = \frac{2 \sigma}{\pi \left(x^2+\sigma^2\right)}$

$\sigma > 0, x > 0$

For an easier to read version of these PDF equations, see the userguide vignette.

References

Denwood, M.J. 2016. runjags: An R Package Providing Interface Utilities, Model Templates, Parallel Computing Methods and Additional Distributions for MCMC Models in JAGS. J. Stat. Softw. 71. doi:10.18637/jss.v071.i09.

See Also

load.pkg.module

Examples

library("rjags")
# The JAGS module is loaded when the package is attached:
list.modules()
# But can be unloaded:
unload_JAGSmodule()
list.modules()
# And re-loaded:
load_JAGSmodule()
list.modules()

# A simple model using the package module:
m <- "model{ L ~ dlomax(1,1) }"
jags.model(textConnection(m))

---

Load the internal JAGS module provided by runjags

Description

Some basic facilities to test the distributions provided by the package module. TODO: remove and replace with d/p/q/rlomax etc

Usage

test_JAGSmodule(
  distribution,
  funtype,
  parameters,
  x,
  uselog = FALSE,
  lower = TRUE
)

Arguments

distribution	One of lomax, mouch, genpar of halfcauchy
funtype	One of d, p or q
parameters	a vector of parameters
x	the x (for funtype="d"), q (for funtype="p"), or p (for funtype="q") value to use in the distribution
uselog	should the probability be returned (or given) on the log axis?
lower	ignored for funtype="d"

References

Denwood, M.J. 2016. runjags: An R Package Providing Interface Utilities, Model Templates, Parallel Computing Methods and Additional Distributions for MCMC Models in JAGS. J. Stat. Softw. 71. doi:10.18637/jss.v071.i09.

See Also

load_JAGSmodule

Examples

test_JAGSmodule("lomax", "d", c(1,1), 1)
test_JAGSmodule("mouch", "p", c(1), 1)

---