LaLonde

This data set gives the outcomes a well as treatment assignments and covariates for the econometric evaluation of training programs in LaLonde (1986).  Taken from package CBPS.

datasets

Causal effect estimation in observational data often requires identifying a set of untreated observations that are comparable to some treated group of interest.  This package provides a suite of functions for identifying such a set of observations and for implementing standard and new diagnostics tools.  The primary function, svmmatch(), uses support vector machines to identify a region of common support between treatment and control groups.  A sensitivity analysis, balance checking, and assessment of the region of overlap between treated and control groups is included.  The Bayesian implementation allows for recovery of uncertainty estimates for the treatment effect and all other parameters.

Marc Ratkovic

SVMMatch

Causal Effect Estimation and Diagnostics with Support Vector
Machines

LaLonde function

A data frame consisting of 5 columns (including a treatment assignment vector) and 2787 observations.

Format

Data from the National Supported Work Study.  A benchmark matching dataset.  Columns consist of an indicator for whether the observed unit was in the experimental subset; an indicator for whether the individual received the treatment; age in years; schooling in years; indicators for black and Hispanic; an indicator for marriage status, one of married; an indicator for no high school degree; and reported earnings in 1974, 1975, and 1978.  1974 and 1975 earnings are pre-treatment.  1978 earnings is taken as the outcome variable.  Note: This data and description were originally prepared for the CBPS package.

LaLonde: LaLonde Data for Covariate Balancing Propensity Score

Description

Usage

Arguments

Format

Source

References