absstddif: Absolute standardized differences in means.

Description Usage Arguments Value Author(s) References Examples

View source: R/absstddif.r

Description

Function for calculating absolute differences in means between the covariates in the treatment and control groups in terms of the original units of the covariates. Here, the absolute differences in means are normalized by the simple average of the treated and control standard deviations before matching (see Rosenbaum and Rubin 1985 for details).

Usage

1
	absstddif(X_mat, t_ind, std_dif)

Arguments

X_mat

matrix of covariates: a matrix of covariates used to build the rank-based Mahalanobis distance matrix.

t_ind

treatment indicator: a vector of zeros and ones indicating treatment (1 = treated; 0 = control).

std_dif

standardized differences: a scalar determining the number of absolute standardized differences.

Value

A vector that can be used with the mom, near and far options of bmatch and nmatch.

Author(s)

Jose R. Zubizarreta <[email protected]>, Cinar Kilcioglu <[email protected]>.

References

Rosenbaum, P. R., and Rubin, D. B. (1985), "Constructing a Control Group by Multivariate Matched Sampling Methods that Incorporate the Propensity Score," The American Statistician, 39, 33-38.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
# Load and attach data
data(lalonde)
attach(lalonde)

# Treatment indicator
t_ind = treatment

# Constrain differences in means to be at most .05 standard deviations apart
mom_covs = cbind(age, education, black, hispanic, married, nodegree, re74, re75)
mom_tols = absstddif(mom_covs, t_ind, .05)

Example output

Loading required package: lattice
Loading required package: MASS
Loading required package: slam
Loading required package: Rglpk
Using the GLPK callable library version 4.52

designmatch documentation built on June 19, 2018, 1 a.m.