Skip to contents

Simple Cluster Based Meat Matrix Estimator

Source: R/methods.R
meatHC.Rd

This (simplified) method for a new S3 generic based on meatHC computes the meat matrix for a fitted glmmTMB model, which is the cross-product of the cluster-wise score vectors (empirical estimating functions) extracted by estfun.

Usage

meatHC(x, ...)

# Default S3 method
meatHC(x, ...)

# S3 method for class 'glmmTMB'
meatHC(x, ...)

Arguments

x

a glmmTMB object fitted with ML (REML is not supported).

...

additional arguments passed to estfun, in particular full, cluster and rawnames arguments.

Value

A square matrix where each element represents the cross-product of the score vectors for the parameters in the model. The rows and columns are named according to the parameter names.

Note

This meat matrix is not scaled by the number of clusters.

Examples

m <- glmmTMB(count ~ mined + (1 | spp), data = Salamanders, family = nbinom1)
meatHC(m)
#>             (Intercept)  minedno
#> (Intercept)    18.45719 18.72652
#> minedno        18.72652 98.85238
meatHC(m, full = TRUE)
#>                  (Intercept)   minedno disp~(Intercept) theta_1|spp.1
#> (Intercept)        18.457190  18.72652         8.767119     -8.880672
#> minedno            18.726521  98.85238        25.052327    -10.824946
#> disp~(Intercept)    8.767119  25.05233       104.156130      5.405059
#> theta_1|spp.1      -8.880672 -10.82495         5.405059      7.825683