dr.cate.regress: Internal function.

In bonander/CEAforests: Tree-based heterogenous cost-effectiveness and policy learning with individual-level data (beta)

Description

dr.cate.regress Fits a linear regression to the estimated heterogeneous effects to assess the determinants of heterogeneity. Uses augmented inverse probability weighting (AIPW) to de-bias the scores, with sandwich standard errors.

Usage

dr.cate.regress(
  dr.scores,
  X = NULL,
  alpha = 0.05,
  robust.se = TRUE,
  subset = NULL
)

Arguments

dr.scores	Doubly robust scores for tau.
X	The variables to include in the model. If NULL, and intercept only model is run.
alpha	The desired significance level, defaults to 0.05.
robust.se	Whether or not to compute robust (sandwich) standard errors. Defaults to TRUE.
subset	A specified subset.

Value

Returns the results from linear regression on the estimated heterogeneous effects with supplied covariates with asymptotically valid standard errors. To be used with causal_forest object directly. For cea_forests or nmb_forests, use the explain_forest() function.

Examples

## Not run: 
To be added...

## End(Not run)

bonander/CEAforests documentation built on April 1, 2021, 10:57 a.m.