Title: Discrete Distribution Approximations
Version: 1.0.3
Description: Creates discretised versions of continuous distribution functions by mapping continuous values to an underlying discrete grid, based on a (uniform) frequency of discretisation, a valid discretisation point, and an integration range. For a review of discretisation methods, see Chakraborty (2015) <doi:10.1186/s40488-015-0028-6>.
License: MIT + file LICENSE
LazyData: true
URL: https://github.com/reconhub/distcrete
BugReports: https://github.com/reconhub/distcrete/issues
Suggests: knitr, rmarkdown, testthat
RoxygenNote: 6.0.1
VignetteBuilder: knitr
NeedsCompilation: no
Packaged: 2017-11-23 12:44:36 UTC; steph
Author: Steph Locke [cre], Rich FitzJohn [aut], Anne Cori [aut], Thibaut Jombart [aut]
Maintainer: Steph Locke <steph@itsalocke.com>
Repository: CRAN
Date/Publication: 2017-11-23 13:50:05 UTC

Discretise a distribution

Description

Discretise a distribution.

Usage

distcrete(name, interval, ..., w = 0.5, anchor = 0)

Arguments

name

The name of a distribution function (e.g., norm, gamma). The distribution must have a cdf function (e.g., pnorm) and a quantile function (e.g., qnorm) defined.

interval

The interval to discretise the interval onto.

...

Parameters to cdf. Can be matched positionally or by name.

w

How to weight the endpoints; must be between 0 and 1. If 0.5 then integration happens centred around the interval, if 0 floor, if 1 then ceiling.

anchor

Any location that is a valid x

Author(s)

Rich FitzJohn

Examples

library(distcrete)
set.seed(415)
d0 <- distcrete("gamma", 1, shape = 3, w = 0)
d0$d(1:10)
d0$p(c(.1,.5))
d0$q(c(.1,.5))
d0$r(10)