傾向性評(píng)分中的結(jié)局變量不用管福扬，其實(shí)沒有用到，根據(jù)因變量調(diào)整所有的協(xié)變量就可以了间驮」幔卡鉗值用來(lái)再次對(duì)沒有匹配的指標(biāo)進(jìn)行進(jìn)一步的調(diào)整。
library(MatchIt)
library(tableone)
rt1=read.table("genesymbol.txt",sep="\t",header=T,check.names=F)
data(lalonde)
head(lalonde,4)
str(lalonde)
dput(names(lalonde))
preBL <- CreateTableOne(vars=c("treat","age","educ","black","hispan","married","nodegree","re74","re75","re78"),
strata="treat",data=lalonde,
factorVars=c("treat","black","hispan","married","nodegree"))
# treat是感興趣變量,re78為結(jié)局變量
print(preBL,showAllLevels = TRUE)
f=matchit(treat~re74+re75+educ+age+married+nodegree,data=lalonde,method="nearest",ratio = 1)
# treat是感興趣變量,re78為結(jié)局變量
summary(f)
matchdata=match.data(f)
mBL <- CreateTableOne(vars=c("treat","age","educ","black","hispan","married","nodegree","re74","re75","re78"),
strata="treat",data=matchdata,
factorVars=c("treat","black","hispan","married","nodegree"))
print(mBL,showAllLevels = TRUE)
plot(f, type = 'jitter', interactive = FALSE)
# hispan不平衡竞帽，需要卡鉗值
f1=matchit(treat~re74+re75+educ+black+hispan+age+married+nodegree,data=lalonde,method="nearest",caliper=0.05)
summary(f1)
matchdata1=match.data(f1)
mBL1 <- CreateTableOne(vars=c("treat","age","educ","black","hispan","married","nodegree","re74","re75","re78"),
strata="treat",data=matchdata1,
factorVars=c("treat","black","hispan","married","nodegree"))
print(mBL1,showAllLevels = TRUE)
plot(f1, type = 'jitter', interactive = FALSE)
#導(dǎo)出數(shù)據(jù)
library(foreign)
matchdata$id<-1:nrow(matchdata)
write.csv(matchdata1,"matchdata.csv")
