A function uses Bayesian methods to incorporate uncertainties in estimated propensity scores and provide adjusted standard errors for propensity score regressions.
Usage
bpsr(Y, t, X, K = 10000, S = 1000)
Arguments
Y
A vector containing the outcome variable.
t
A vector containing the treatment indicator.
X
A matrix containing the covariates.
K
Numbers of iterations.
S
Number of posterior samples.
Details
Estimated propensity scores are used as an additional covariate in the main outcome model to control for selection or to provide better control for the nonlinear effects of covariates. The function bpsr takes into account the uncertainties in estimating the propensity scores and adjusts the standard errors accordingly.
Value
estimates
The estimated treatment effects and their adjusted standard errors. Phat shows the results with unadjusted standard errors. BPSR shows the results with adjusted standard errors using the Bayesian method.
time
The time elapsed for the computation.
sims
The number of iterations requested for the Bayesian computation, K
posterior
The posterior sample distribution of the treatment effects. The function bpsr uses the posterior standard deviation to approximate the standard error.
Weihua An, Huizi Xu, and Zhida Zheng, Indiana University Bloomington.
References
An, Weihua. 2010. "Bayesian Propensity Score Estimators: Incorporating Uncertainties In
Propensity Scores Into Causal Inference." Sociological Methodology 40: 151-189. http://mypage.iu.edu/~weihuaan/Documents/2010_BPSE.pdf.
See Also
bpsm, modelpsm, modelpsr, Match, sortps
Examples
## Not run:
library(boot)
a = 2; b = c(1, -2, 5); N = 500
x1 <- runif(N, 0, 1)
x2 <- runif(N, 0, 1)
X <- cbind(rep(1, N), x1, x2)
p <- inv.logit( X %*% b )
t <- rbinom(N, 1, p)
e <- rnorm(N, 0, 1)
Y <- 5 * t + a * ( x1 + x2 ) + e
bpsr(Y, t, X )
## End(Not run)