est_tsm: Efficient Estimation of Counterfactual Means of Dynamic Rules
In Netflix/sherlock: Causal Machine Learning for Segment Discovery and Analysis

Description Usage Arguments Value

Evaluate the treatment-specific mean (TSM) based on static interventions or on the dynamic treatment rule. Efficient one-step and TML estimators are available for each of these target parameters.

1	est_tsm(data_with_rule, tsm_param, est_type)

`data_with_rule`	A `data.table` containing the input data, augmented with cross-validated nuisance parameter estimates, an estimate of the CATE, and a treatment rule assigned based on the estimated CATE via `assign_rule`. This input object should be created by successive calls to `set_est_data`, `est_cate`, and `assign_rule` in a sequence, or through a wrapper function that composes these function calls automatically.
`tsm_param`	A `character` string (of length one) identifying the treatment-specific mean to be estimated. `"tsm_static_trt"` gives the counterfactual mean under the static intervention assigning treatment to all units; `"tsm_static_ctl"` gives the counterfactual mean under the static intervention withholding treatment from all units; and, lastly, `"tsm_dynamic_opt"` gives the counterfactual mean under the dynamic rule assigning treatment based on potential benefit/harm from treatment.
`est_type`	Specification of either the one-step or TML estimator. See the documentation of `est_effect` for details.

A data.table containing point estimates and the standard error of the estimates of the specified target parameter. The resulting object contains the estimated efficient influence function as an attribute appended to the object via setattr.

Netflix/sherlock documentation built on Dec. 17, 2021, 5:22 a.m.