mixture_score_glm: Create score-based mixture via glm

In ck37/tlmixture: Data-adaptive Creation Of Exposure (Treatment) Mixtures Using Targeted Learning

Description

Create score-based mixture via glm

Usage

mixture_score_glm(
  data,
  outcome,
  exposures,
  estimator_mixture = sl3::Lrnr_sl$new(learners = sl3::make_learner(sl3::Stack,
    sl3::make_learner(sl3::Lrnr_mean), sl3::make_learner(sl3::Lrnr_glm)), metalearner =
    sl3::make_learner(sl3::Lrnr_nnls)),
  estimator_confounders = NULL,
  folds_mixture = 5L,
  folds_confounders = 5L,
  exposure_groups = NULL,
  max_iterations = 5L,
  tolerance = 1e-05,
  adjust_other_exposures = TRUE,
  force_offset = TRUE,
  quantiles = NULL,
  family = "continuous",
  verbose = FALSE,
  debug = FALSE,
  ...
)

Arguments

data	tbd
outcome	outcome column name
exposures	tbd
estimator_mixture	sl3::Lrnr_sl with offset-supporting learners
estimator_confounders	sl3::Lrnr_sl with offset-supporting learners
folds_mixture	SL folds to use during mixture estimation; default 5. (Not yet implemented).
folds_confounders	SL folds to use during confounder estimation; default 5. (Not yet implemented).
exposure_groups	List of all exposure groups
max_iterations	tbd
tolerance	tbd
adjust_other_exposures	Whether or not to include other exposure groups as confounders to adjust for. Default TRUE.
force_offset	Whether to use offset style estimation for gaussian outcomes; default True.
quantiles	tbd
family	Not used for this mixture estimator.
verbose	tbd
debug	If True show very detailed output.
...	tbd

ck37/tlmixture documentation built on Dec. 19, 2021, 5:13 p.m.