Kang_Schafer_Simulation: Kang and Schafer simulation

In fiona19832008/PSLB: Propensity Score Analysis with Local Balance

Kang and Schafer simulation

Description

This function generate the simulation scenarios presented in Kang and Schafer (2007)

Usage

Kang_Schafer_Simulation(
  n,
  beta = c(-1, 0.5, -0.25, -0.1),
  alpha = c(210, 27.4, 13.7, 13.7, 13.7),
  mu = rep(0, 4),
  sd = diag(4),
  seeds
)

Arguments

n	The number of sample size.
beta	The coefficient of the true propensity score model. Default to c(-1,0.5,-0.25,-0.1).
alpha	The coefficient of the true outcome model. Defaults to c(210,27.4,13.7,13.7,13.7).
mu	The mean of the covariates Z in the true propensity score model. Default to rep(0,4)
sd	The variance of the covariates Z in the true propensity score model. Default to diag(4)
seeds	The seed number

Value

A list containing the following components:

• "Data": The simulated data matrix includes the outcome Y (1st column), the treatment assignment Tr (2nd column), the covariates Z in the true propensit score mode (3rd to 6th column), the observed covariates X (7th to 10th column), and the true propensity score PS (11th column).

• "Treat.effect": The mean of the outcome Y

References

Kang, J. D. and Shafer, J. L. (2007) Demystifying double robustness: A comparison of alternative strategies for estimating a population mean from incomplete data. Statistical Science, 22, 523-539.

fiona19832008/PSLB documentation built on April 14, 2022, 12:41 a.m.