nikola-sur / goodGLM

A generalized Hosmer-Lemeshow goodness-of-fit test for a family of generalized linear models (GLMs)

A generalized Hosmer-Lemeshow (GHL) goodness-of-fit test for a family of generalized linear models (GLMs).

Installation

To install this package, type:

# install.packages('devtools')
devtools::install_github('https://github.com/nikola-sur/goodGLM/')

Example usage

First, we prepare some artificial data.

n <- 100L
b0 <- 1.0
b1 <- 0.1

x <- runif(n, 0, 1)
mu <- exp(b0 + b1*x)
y <- rpois(n, lambda = mu)

We then fit a GLM and perform the goodness-of-fit test!

mod <- glm(y ~ x, family = poisson(link = 'log'))

library(goodGLM)
gof_output <- goodGLM(mod, groups = 10L, group_mode = "variance")
gof_output

Citation

If you use this package, please consider citing the package

@software{goodGLM_2022,
author = {Nikola, Surjanovic and Richard, Lockhart and Thomas, Loughin},
month = {2},
title = {{'goodGLM': A generalized Hosmer-Lemeshow goodness-of-fit test for a family of generalized linear models}},
url = {https://github.com/nikola-sur/goodGLM},
version = {0.0.0.9},
year = {2022}
}

and/or one of our two papers on the GHL test:

@article{surjanovic2020generalized,
  title={A Generalized Hosmer-Lemeshow Goodness-of-Fit Test for a Family of Generalized Linear Models},
  author={Surjanovic, Nikola and Lockhart, Richard and Loughin, Thomas M},
  journal={arXiv preprint arXiv:2007.11049},
  year={2020}
}

@article{surjanovic2021improving,
  title={Improving the Hosmer-Lemeshow Goodness-of-Fit Test in Large Models with Replicated Trials},
  author={Surjanovic, Nikola and Loughin, Thomas M},
  journal={arXiv preprint arXiv:2102.12698},
  year={2021}
}