Fit Negative Binomial Mixed Model via PQL

Fit negative binomial mixed model (GLMM) with a single random effect using penalized quasi-likelihood (PQL)

Usage

fastglmm.nb(
  formula,
  data,
  weights = NULL,
  maxit = 100,
  tol = 0.001,
  tol.eta = 0.001,
  doCoxReid = nrow(data) < 1000,
  nthreads = 6
)

Arguments

formula

a two-sided linear formula object describing both the fixed-effects and random-effects part of the model, with the response on the left of a ~ operator and the terms, separated by + operators, on the right. Random-effects terms are distinguished by vertical bars (|) separating expressions for design matrices from grouping factors.

data

an optional data frame containing the variables named in

weights

an optional vector of prior weights with a value for each sample.

maxit

max number of NB iterations

tol

convergence criterion for the 1D search of the delta space

tol.eta

convergence criterion eta in the PQL iteration

doCoxReid

use Cox-Reid correction for estimating theta in negative binomial model

nthreads

number of threads

Examples

library(MASS)
data(PsychAD)

# regression formula
form <- PTPRG ~ (1|SubID) + offset(log(libSize))

# NB GLMM on PTPRG expression via PQL
fit1 <- fastglmm.nb(form, PsychAD)
coef(summary(fit1))
#>              Estimate Std. Error    df   t value      Pr(>|t|)
#> (Intercept) -8.638731 0.04585474 148.2 -188.3934 2.647753e-178

# NB GLMM via Laplace approximation
# fit2 <- lme4::glmer.nb(form, PsychAD)
# coef(summary(fit2))