Extracts Random Effects from Functional Columns
Source:R/PipeOpFDARandomEffect.R
mlr_pipeops_fda.random_effect.Rd
This is the class that extracts random effects, specifically random intercepts and
random slopes, from functional columns. This PipeOp fits a linear mixed model, specifically
a random intercept and random slope model, using the lme4::lmer()
function.
The target variable is the value of the functional feature which is regressed on the functional feature's argument
while subject id determines the grouping structure. After model estimation, the random effects are extracted and
assigned to the correct id.
Parameters
The parameters are the parameters inherited from PipeOpTaskPreprocSimple
.
Naming
The new names append _random_intercept
and _random_slope
to the corresponding column name of the
functional feature.
Super classes
mlr3pipelines::PipeOp
-> mlr3pipelines::PipeOpTaskPreproc
-> mlr3pipelines::PipeOpTaskPreprocSimple
-> PipeOpFDARandomEffect
Methods
Method new()
Initializes a new instance of this Class
Usage
PipeOpFDARandomEffect$new(id = "fda.random_effect", param_vals = list())
Arguments
id
(
character(1)
) Identifier of the operator, default is"fda.random_effect"
.param_vals
(named
list()
) List of hyperparameter settings, overwriting default settings set during construction.