Provides probability mass, distribution, quantile, random-variate generation, and method-of-moments parameter-estimation functions for the Delaporte distribution. The Delaporte is a discrete probability distribution which can be considered the convolution of a negative binomial distribution with a Poisson distribution. Alternatively, it can be considered a counting distribution with both Poisson and negative binomial components. It has been studied in actuarial science as a frequency distribution which has more variability than the Poisson, but less than the negative binomial.
Delaporte is an
R package which provides the probability mass, distribution, quantile, random variate generation, and method of moments parameter estimation functions for the Delaporte distribution. As the distribution does not have a closed form but requires summations or double summations to calculate values, the functions have been programmed in Fortran and C. In cases where approximations are sufficient, the quantile and random variate generator have the option to use a much faster Poisson-negative binomial estimate as opposed to the full Delaporte double summations.
The author is grateful to Drew Schmidt both generally for his writings on R, C++, and Fortran and specifically for help with this project.
Please ensure that all contributions comply with both R and CRAN standards for packages.
This project attempts to follow Semantic Versioning
This project attempts to follow the changelog system at Keep a CHANGELOG
This project intends to have as few dependancies as possible. Please consider that when writing code.
Please review and conform to the current code stylistic choices (e.g. 80 character lines, two-space indentations).
Please provide valid .Rd files and not roxygen-style documentation.
Please review the current test suite and supply similar
testthat-compatible unit tests for all added functionality.
If you would like to contribute to the project, it may be prudent to first contact the maintainer via email. A request or suggestion may be raised as an issue as well. To supply a pull request (PR), please:
At this point, the PR will be discussed and eventually accepted or rejected.