aipw_ate: Estimate average treatment effect (ATE) with augmented...

In jackcollison/causality: Tools for Performing Causal Inference with Observational Data

Description

Estimate average treatment effect (ATE) with augmented inverse propensity score weighting.

Usage

aipw_ate(data, x, y, w, p, cf = TRUE)

Arguments

data	a dataframe object containing the variables and values.
x	a list of character vectors specifying variables to be included in the model (columns in the data). If unspecified, then it is assumed to be all columns in the data besides y and w.
y	a character vector specifying the response variable.
w	a character vector specifying the treatment status.
p	a vector containing propensity score values.
cf	logical; if TRUE then includes confidence interval on ATE.

Details

Estimates the ATE τ using the doubly robust method described in Scharfstein, Robins and Rotznitzky (1998) that combines both regression and propensity score weighting.

τ = E ≤ft[ W_i \frac{Y_i-τ(1,X_i)}{e(X_i)} + (1-W_i) \frac{Y_i-τ(0,X_i)}{1-e(X_i)} + τ(1,X_i) - τ(0,X_i)\right]

where e(X_i) = P(W_i = 1 | X_i = x) is the propensity score and τ(1, X_i) and τ{0, X_i} are estimated in the first stage via OLS regression.

Value

a list of ATE, 95 percent confidence interval upperbound and lowerbound or just ATE, depending on user input of cf.

References

Rotnitzky, Andrea, James M. Robins, and Daniel O. Scharfstein. 1998. “Semiparametric Regression for Repeated Outcomes with Nonignorable Nonresponse." Journal of the American Statistical Association. Vol. 93, No. 444, Dec. pgs. 1321-1339.

Cao, Weihua, Anastasios A. Tsiatis, and Marie Davidian. 2009. “Improving efficiency and robustness of the doubly robust estimator for a population mean with incomplete data". Biometrika. Vol. 96(3). pgs. 723–734.

Examples

data("lalonde")

# calculate propensity scores
p <- propensity_score(lalonde, y = "re78", w = "treat", model = "logistic")
ate <- aipw_ate(data = lalonde, y = "re78", w = "treat", p = p)

jackcollison/causality documentation built on Dec. 20, 2021, 8:05 p.m.