change_complexity: Change the complexity of a fit_estimate_partition

In microsoft/CausalGrid: Analysis of Subgroups

Change the complexity of a fit_estimate_partition

Description

Change the complexity level of the partition and re-estimate cell statistics. If you have a minimal estimated_partition then you need to pass in the other params.

Usage

change_complexity(
  fit,
  y,
  X,
  d = NULL,
  partition_i,
  index_tr = fit$index_tr,
  split_seq = fit$split_seq,
  est_plan = fit$est_plan
)

Arguments

fit	estimated_partition
y	Nx1 matrix of outcome (label/target) data. With multiple core estimates see Details below.
X	NxK matrix of features (covariates). With multiple core estimates see Details below.
d	(Optional) NxP matrix (with colnames) of treatment data. If all equally important they should be normalized to have the same variance. With multiple core estimates see Details below.
partition_i	partition_i - 1 is the last include in split_seq included in new partition
index_tr	Split between train and estimate samples (default is to get from fit)
split_seq	sequential list of splits (default is to get from fit)
est_plan	EstimatorPlan.

Details

Note: doesn't update the importance weights

Value

updated estimated_partition

Multiple estimates

With multiple core estimates (M) there are 3 options (the first two have the same sample across treatment effects).

1. DS.MULTI_SAMPLE: Multiple pairs of (Y_m,W_m). y,X,d are then lists of length M. Each element then has the typical size The N_m may differ across m. The number of columns of X will be the same across m.

2. DS.MULTI_D: Multiple treatments and a single outcome. d is then a NxM matrix.

3. DS.MULTI_Y: A single treatment and multiple outcomes. y is then a NXM matrix.

microsoft/CausalGrid documentation built on Aug. 25, 2022, 9:30 a.m.