EnsemTree: Ensemble Tree (ET)

In ellenxtan/ifedtree: Tree-based Federated Learning Approach for Personalized Effect Estimation and Prediction

Ensemble Tree (ET)

Description

Build and/or predict on an ensemble regression tree.

Usage

EnsemTree(
  coord_id,
  aug_df,
  site,
  covars,
  honest = FALSE,
  is_pred = FALSE,
  myfit = NULL,
  leaves_tab = NULL,
  ...
)

Arguments

coord_id	Site index for coordinating site.
aug_df	The augmented data frame used to fit an ensemble tree ('data.table').
site	Variable name for site indicator.
covars	A vector of covariate names used.
honest	Whether to use honest splitting (i.e., subsample splitting). Default is FALSE.
is_pred	Whether to build an ensemble tree or make prediction. Default is FALSE.
myfit	A fitted ensemble tree (for prediction purpose). Default is NULL.
leaves_tab	A table indicates leaf assignments in a fitted ensemble tree (for prediction purpose). Default is NULL.
\dots	Additional arguments for building the tree.

Value

Training: return a fitted ensemble tree and estimation of the input data; Prediction: return predictions of the input data.

Examples

data(SimDataLst)
K <- length(SimDataLst)
covars <- grep("^X", names(SimDataLst[[1]]), value=TRUE)
fit_lst <- list()
for (k in 1:K) {
    tmpdf <- SimDataLst[[k]]
    # use your estimator of interest
    fit_lst[[k]] <- grf::causal_forest(X=as.matrix(tmpdf[, covars, with=FALSE]),
                                       Y=tmpdf$Y, W=tmpdf$Z)
}

coord_id <- 1
coord_test <- GenSimData(coord_id)

coord_df <- SimDataLst[[coord_id]]
aug_df <- GenAugData(coord_id, coord_df, fit_lst, covars)

res_et <- EnsemTree(coord_id, aug_df, "site", covars)
et_hat <- EnsemTree(coord_id, coord_test, "site", covars, is_pred=TRUE,
          myfit=res_et$myfit, leaves_tab=res_et$leaves_tab)

res_et <- EnsemTree(coord_id, aug_df, "site", covars, honest=TRUE)
et_hat <- EnsemTree(coord_id, coord_test, "site", covars, honest=TRUE, is_pred=TRUE,
          myfit=res_et$myfit, leaves_tab=res_et$leaves_tab)

ellenxtan/ifedtree documentation built on March 28, 2023, 9:09 a.m.