DR: Doubly robust estimator

In youmisuk/CURobustML: Robust Machine Learning Under Cluster-Level Unmeasured Confounding

Doubly robust estimator

Description

Doubly robust estimator

Usage

DR(Y, Z, interZ = formula(~1), Z.hat, Y1.hat, Y0.hat, data)

Arguments

Y	continuous outcome variable
Z	binary treatment indicator, 1 - treatment, 0 - control
interZ	formula that contains the variables that "interact" with the treatment. "1" will be always added. The default is no interaction, i.e., formula = formula(~1).
Z.hat	treatment/propensity score prediction from DRPRcomb
Y1.hat	outcome prediction among treated units from DRPRcomb
Y0.hat	outcome prediction among untreated units from DRPRcomb
data	dataframe containing the variables in the model

Value

estimates and standard errors

Examples

DRPRcomb.rslt <- DRPRcomb(Y=twolevel_data$Y, Z=twolevel_data$Z, interZ=(~ W1),
 X=twolevel_data[, c("X1", "X2", "X3", "W1")], ID=twolevel_data$id,
 library="glm", data=twolevel_data) # the default algorithms are "glm" and "deeplearning".

# with final predictions
DR(Y=twolevel_data$Y, Z=twolevel_data$Z, interZ=(~ W1), Z.hat=DRPRcomb.rslt$Z.hat,
 Y1.hat=DRPRcomb.rslt$Y1.hat, Y0.hat=DRPRcomb.rslt$Y0.hat, data=twolevel_data)

# with predictions from glm with fixed effects of clusters
DR(Y=twolevel_data$Y, Z=twolevel_data$Z, interZ=(~ W1),
 Z.hat=DRPRcomb.rslt$Z.hats$Ztest.hat_glm_1, Y1.hat=DRPRcomb.rslt$Y1.hats$Y1test.hat_glm_1,
 Y0.hat=DRPRcomb.rslt$Y0.hats$Y0test.hat_glm_1, data=twolevel_data)

# with predictions from glm wtih random effects of clusters
DR(Y=twolevel_data$Y, Z=twolevel_data$Z, interZ=(~ W1),
 Z.hat=DRPRcomb.rslt$Z.hats$Ztest.hat_glm_3, Y1.hat=DRPRcomb.rslt$Y1.hats$Y1test.hat_glm_3,
 Y0.hat=DRPRcomb.rslt$Y0.hats$Y0test.hat_glm_3, data=twolevel_data)

youmisuk/CURobustML documentation built on Sept. 11, 2022, 11:04 a.m.