#=====================================================
#code for Chapter 5 "Propensity Score Matching" of book:
#Leite, W. L. (2017). Practical propensity score methods using R. 
#Thousand Oaks, CA: Sage. 

#PART 2 - ONE TO ONE GREEDY MATCHING WITH REPLACEMENT

# 
#this is the code that was used to generate the example results in the book
#As the R software and the R packages used in this example are updated frequently
#some incompatibilities between the current code and new R versions or package versions
#may appear
#Any updates to the code will be posted at:
# http://www.practicalpropensityscore.com


#This example estimates the effect of mothers having a job 
#that provides or subsidizes child care 
#on the length that they breastfeed their children 
#National Longitudinal Survey of Youth 1979 (NLSY79) 
#and the NLSY79 Children and Youth

#Perform One-to-One greedy matching with replacement and 
#with caliper to estimate the ATT

#load data
load(file="Chapter5_data_with_propensity_scores_and_formula.rData")


library(MatchIt) #library for propensity score matching
greedyMatching <- matchit(psFormula,distance=data$logitPScores, 
                     data = data, 
                     method = "nearest",ratio=1,replace=T,caliper=0.25)


#diagnose covariate balance
(balance.greedyMatching <- summary(greedyMatching, standardize=T))

#obtain the summary of balance aftdr matching 
summary(abs(balance.greedyMatching$sum.matched$"Std. Mean Diff."))
table(abs(balance.greedyMatching$sum.matched$"Std. Mean Diff.") > 0.1)

#estimate ATT with one-to-one greedy matching with replacement
#within 0.25 caliper

#obtain matched data
data.greedyMatching <- match.data(greedyMatching)



library(survey)
design.greedyMatching <- svydesign(ids=~1, weights=~weights,
                                   data=data.greedyMatching)
#estimate the ATT 
model.greedyMatching <- svyglm(C0338600~childCare, design.greedyMatching, family=gaussian())
summary(model.greedyMatching)