glmmTMB (version 1.1.7)

Anova.glmmTMB: Downstream methods

Description

Methods have been written that allow glmmTMB objects to be used with several downstream packages that enable different forms of inference. For some methods (Anova and emmeans, but not effects at present), set the component argument to "cond" (conditional, the default), "zi" (zero-inflation) or "disp" (dispersion) in order to produce results for the corresponding part of a glmmTMB model. Support for emmeans also allows additional options component = "response" (response means taking both the cond and zi components into account), and component = "cmean" (mean of the [possibly truncated] conditional distribution).

In particular,

• car::Anova constructs type-II and type-III Anova tables for the fixed effect parameters of any component

• the emmeans package computes estimated marginal means (previously known as least-squares means) for the fixed effects of any component, or predictions with type = "response" or type = "component". Note: In hurdle models, component = "cmean" produces means of the truncated conditional distribution, while component = "cond", type = "response" produces means of the untruncated conditional distribution.

• the effects package computes graphical tabular effect displays (only for the fixed effects of the conditional component)

Usage

Anova.glmmTMB(
  mod,
  type = c("II", "III", 2, 3),
  test.statistic = c("Chisq", "F"),
  component = "cond",
  vcov. = vcov(mod)[[component]],
  singular.ok,
  ...
)
Effect.glmmTMB(focal.predictors, mod, ...)

Arguments

mod

a glmmTMB model

type

type of test, "II", "III", 2, or 3. Roman numerals are equivalent to the corresponding Arabic numerals. See Anova for details.

test.statistic

unused: only valid choice is "Chisq" (i.e., Wald chi-squared test)

component

which component of the model to test/analyze ("cond", "zi", or "disp") or, in emmeans only, "response" or "cmean" as described in Details.

vcov.

variance-covariance matrix (usually extracted automatically)

singular.ok

OK to do ANOVA with singular models (unused) ?

...

Additional parameters that may be supported by the method.

focal.predictors

a character vector of one or more predictors in the model in any order.

Details

While the examples below are disabled for earlier versions of R, they may still work; it may be necessary to refer to private versions of methods, e.g. glmmTMB:::Anova.glmmTMB(model, ...).

Examples

warp.lm <- glmmTMB(breaks ~ wool * tension, data = warpbreaks)
salamander1 <- up2date(readRDS(system.file("example_files","salamander1.rds",package="glmmTMB")))
if (require(emmeans)) withAutoprint({
    emmeans(warp.lm, poly ~ tension | wool)
    emmeans(salamander1, ~ mined, type="response")  # conditional means
    emmeans(salamander1, ~ mined, component="cmean")     # same as above, but re-gridded
    emmeans(salamander1, ~ mined, component="zi", type="response")  # zero probabilities
    emmeans(salamander1, ~ mined, component="response")  # response means including both components
})
if (getRversion() >= "3.6.0") {
   if (require(car)) withAutoprint({
       Anova(warp.lm,type="III")
       Anova(salamander1)
       Anova(salamander1, component="zi")
   })
   if (require(effects)) withAutoprint({
       plot(allEffects(warp.lm))
       plot(allEffects(salamander1))
   })
}