Skip to contents

Tidy, glance, and augment methods for mixeff fits

Source: R/broom.R
mm_broom.Rd

These implement the broom / broom.mixed generics for mm_lmm and mm_glmm fits, so tidy(), glance(), and augment() work on mixeff models the same way they do on lme4 fits.

Arguments

x, data

A fitted mm_lmm or mm_glmm (and, for augment(), optional data to augment; defaults to the model frame).

effects

Which terms to return: any of "fixed", "ran_pars", "ran_vals".

conf.int

Logical; add Wald conf.low/conf.high for fixed effects.

conf.level

Confidence level for conf.int.

...

Unused; for generic compatibility.

Value

A data frame.

Details

tidy() returns one row per model term. effects = "fixed" yields the fixed-effect coefficients (estimate, std.error, statistic, and, for GLMMs, a Wald p.value); effects = "ran_pars" yields the variance- component standard deviations (sd__<term>), correlations (cor__<a>.<b>), and the residual SD (sd__Observation); effects = "ran_vals" yields the conditional modes. glance() returns a one-row model-summary frame; augment() returns the model frame with .fitted and .resid columns.

These methods are registered with generics when the package is loaded; call them via broom::tidy() / broom.mixed::tidy() etc.