glmBayes: glmBayes: Bayesian Generalized Linear Models

In abnormally-distributed/cvreg: Cross Validation and Robust Estimation Utilities

Description

glmBayes: Bayesian Generalized Linear Models

This implements the Zellner-Siow Cauchy g-prior for generalized linear models.

Usage

glmBayes(
  formula,
  data,
  family = "gaussian",
  link = NULL,
  g = NULL,
  max_iter = 9000,
  sim = FALSE,
  nsim = 4000,
  threshold = 1e-09
)

glmBayes(
  formula,
  data,
  family = "gaussian",
  link = NULL,
  g = NULL,
  max_iter = 9000,
  sim = FALSE,
  nsim = 4000,
  threshold = 1e-09
)

Arguments

formula	model formula
data	a data frame
family	"gaussian", "poisson", "negbinom", "binomial", "gamma", "invgauss"
link	the link function. see details for available options.
g	a custom value for g prior. If left as NULL, the default is the ratio of sample size to predictors.
max_iter	max iterations for IRWLS algoritm
sim	if TRUE, draws the specified number of samples from the posterior distribution. defaults to FALSE.
nsim	the number of posterior samples. defaults to 4000.
threshold	tolerance for convergence
x	argument
...	other arguments

Details

This implements the Zellner-Siow Cauchy g-prior for generalized linear models

The following distributions and corresponding link functions are available:

Gaussian: "identity"
Binomial: "logit", "probit", "cauchit", "robit" (Student T with 3 df), and "cloglog"
Poisson & Negative Binomial: "log"
Gamma: "inverse" (1 / x)
Inverse Gaussian: "invsquare" (1/x^2)

Examples

glmBayes()

abnormally-distributed/cvreg documentation built on May 3, 2020, 3:45 p.m.