fluc_update: Fit Fluctuation Model for Targeted Minimum Loss Estimation
In Netflix/sherlock: Causal Machine Learning for Segment Discovery and Analysis

Fit Fluctuation Model for Targeted Minimum Loss Estimation

1	fluc_update(data_with_rule, aux_covar, tol = 5)

`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.
`aux_covar`	A `character` string (of length one) indicating the column of the input data object, `data_with_rule`, that corresponds to the auxiliary ("clever") covariate of the fluctuation model. Note that the form of this term depends on the efficient influence function of the target parameter and should be computed externally and appended to the input data.
`tol`	A `numeric` indicating the maximum size of the parameter estimate of the one-dimensional fluctuation model. Any estimates larger than this tolerance value are set to zero.

A data.table with all of the information contained in the input but augmented to include additional columns for the updated outcome mechanism based on the one-dimensional fluctuation model.

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