phit0/frequentistkiller: Metropolis-Hastings algorithm using IWLS for selected generalized linear models
Frequentistkiller is a package that allows to fit a multivariate generalized linear model for Bernoulli (link = logit), Normal and Poisson (link = log) distributed response variables, including prior asumptions about the regression parameter. Parameters are sampled from a normal proposal density constructed via IWLS during the iterations of a Metropolis-Hastings algorithm and returned as a chain of close-to independent parameters. Their distribution converges to the true posterior distribution of the parameters for infinite iterations.
Getting started
Browse package contents
Package details |
|
|---|---|
| Maintainer | |
| License | MIT + file LICENSE |
| Version | 0.0.0.9000 |
| URL | https://github.com/phit0/frequentistkiller |
| Package repository | View on GitHub |
| Installation |
Install the latest version of this package by entering the following in R:
|
