Extract Random Effects

Source: R/methods.R

Extract random effects from a fitted glmmTMB model, both for the conditional model and zero inflation.

# S3 method for class 'glmmTMB'
ranef(object, condVar = TRUE, ...)

# S3 method for class 'ranef.glmmTMB'
as.data.frame(x, ...)

# S3 method for class 'glmmTMB'
coef(object, condVar = FALSE, ...)

Arguments

object

a glmmTMB model.

condVar

whether to include conditional variances in result.

...

some methods for this generic function require additional arguments (they are unused here and will trigger an error)

x

a ranef.glmmTMB object (i.e., the result of running ranef on a fitted glmmTMB model)

Value

  • For ranef, an object of class ranef.glmmTMB with two components:

    cond

    a list of data frames, containing random effects for the conditional model.

    zi

    a list of data frames, containing random effects for the zero inflation.

    disp

    a list of data frames, containing random effects for the dispersion model.

    If condVar=TRUE, the individual list elements within the cond, zi, and disp components (corresponding to individual random effects terms) will have associated condVar attributes giving the conditional variances of the random effects values. These are in the form of three-dimensional arrays: see ranef.merMod for details. The only difference between the packages is that the attributes are called ‘postVar’ in lme4, vs. ‘condVar’ in glmmTMB.

  • For coef.glmmTMB: a similar list, but containing the overall coefficient value for each level, i.e., the sum of the fixed effect estimate and the random effect value for that level. Conditional variances are not yet available as an option for coef.glmmTMB.

  • For as.data.frame: a data frame with components

    component

    part of the model to which the random effects apply (conditional or zero-inflation)

    grpvar

    grouping variable

    term

    random-effects term (e.g., intercept or slope)

    grp

    group, or level of the grouping variable

    condval

    value of the conditional mode

    condsd

    conditional standard deviation

Note

When a model has no zero inflation, the ranef and coef print methods simplify the structure shown, by default. To show the full list structure, use print(ranef(model),simplify=FALSE) or the analogous code for coef. In all cases, the full list structure is used to access the data frames, see example.

See also

fixef.glmmTMB.

Examples

if (requireNamespace("lme4")) {
   data(sleepstudy, package="lme4")
   model <- glmmTMB(Reaction ~ Days + (1|Subject), sleepstudy)
   rr <- ranef(model)
   print(rr, simplify=FALSE)
   ## extract Subject conditional modes for conditional model
   rr$cond$Subject
   as.data.frame(rr)
}
#> $cond
#> $cond$Subject
#>     (Intercept)
#> 308   40.635100
#> 309  -77.565868
#> 310  -62.878597
#> 330    4.390390
#> 331   10.178966
#> 332    8.191284
#> 333   16.440371
#> 334   -2.986056
#> 335  -45.117116
#> 337   71.919659
#> 349  -21.119006
#> 350   14.059946
#> 351   -7.833567
#> 352   36.245868
#> 369    7.010745
#> 370   -6.339513
#> 371   -3.282264
#> 372   18.049738
#> 
#> 
#> $zi
#> list()
#> 
#> $disp
#> list()
#> 
#>         component  grpvar        term grp    condval   condsd
#> cond.1       cond Subject (Intercept) 308  40.635100 12.53484
#> cond.2       cond Subject (Intercept) 309 -77.565868 12.65076
#> cond.3       cond Subject (Intercept) 310 -62.878597 12.59611
#> cond.4       cond Subject (Intercept) 330   4.390390 12.49122
#> cond.5       cond Subject (Intercept) 331  10.178966 12.49348
#> cond.6       cond Subject (Intercept) 332   8.191284 12.49251
#> cond.7       cond Subject (Intercept) 333  16.440371 12.49794
#> cond.8       cond Subject (Intercept) 334  -2.986056 12.49095
#> cond.9       cond Subject (Intercept) 335 -45.117116 12.54509
#> cond.10      cond Subject (Intercept) 337  71.919659 12.62843
#> cond.11      cond Subject (Intercept) 349 -21.119006 12.50264
#> cond.12      cond Subject (Intercept) 350  14.059946 12.49600
#> cond.13      cond Subject (Intercept) 351  -7.833567 12.49235
#> cond.14      cond Subject (Intercept) 352  36.245868 12.52583
#> cond.15      cond Subject (Intercept) 369   7.010745 12.49202
#> cond.16      cond Subject (Intercept) 370  -6.339513 12.49178
#> cond.17      cond Subject (Intercept) 371  -3.282264 12.49100
#> cond.18      cond Subject (Intercept) 372  18.049738 12.49943