Software and replication material
R package CMatching: Matching Algorithms for Causal Inference with Clustered Data (with M. Cannas and E. Colicino)
Provides functions to perform matching algorithms for causal inference with clustered data, as described in B. Arpino and M. Cannas (2016) <doi:10.1002/sim.6880>. Pure within-cluster and preferential within-cluster matching are implemented. Both algorithms provide causal estimates with cluster-adjusted estimates of standard errors.
The package is described in an R Journal article.
How to implement propensity score matching with clustered data in Stata (presented at the 2018 Spanish Stata meeting)
I show how to implemented in Stata the methods described in Arpino and Cannas (2016) (see above) starting from the psmatch2
package. Slides ("Arpino Stata 2018.pdf"), dofile ("matching.do") and data ("schools.dta) to replicate the analyses can be found at the bottom of this page.
Code to implement the methods and simulation studies in Arpino and Cannas (2016)
In this document we provide additional material for the paper "Propensity score matching with clustered data. An application to the estimation of the impact of caesarean section on the Apgar score" (Arpino and Cannas, 2016). Section A provides the exact formulas for equations used to generate the true treatment and outcome data in the simulation studies. Section B provides the R code used in the baseline simulations. Section C includes full tables of simulation results not included in the paper for the sake of saving space.
Code to implement the methods and analyses in Arpino, De Cao and Peracchi (2014)
R code to implement the partial identification methods in Arpino, De Cao and Peracchi (2014); dasets to replicate the empirical analyses. Arpino_DeCao_Peracchi2014