丁香园临床预测模型课程第四章

今朝落庭霰

已于 2024-12-31 22:38:52 修改

阅读量138

点赞数 3

CC 4.0 BY-SA版权

分类专栏：丁香园临床预测模型学习笔记文章标签： r语言机器学习

于 2024-12-22 17:33:24 首次发布

本文链接：https://blog.youkuaiyun.com/weixin_54841603/article/details/144649448

丁香园临床预测模型学习笔记专栏收录该内容

8 篇文章

订阅专栏

本章介绍倾向性评分匹配（PSM）

#第4章 倾向性评分匹配PSM ####

#用MatchIt包做PSM
install.packages("MatchIt")
library(MatchIt)
lalonde <- read.dta("D:/BaiduNetdiskDownload/lalonde.dta")
#数据集lalonde变量解释：treat是分组变量，age年龄，educ教育程度，black人种，hispan是否是西班牙人
#married是否已婚，nodegree是否有学位，re74和re75基础工资，re78结局变量
head(lalonde)
f <- matchit(treat ~ age + educ + black + hispan + 
               married + nodegree + re74 + re75,data = lalonde,method = "nearest")
summary(f)
f1 <- matchit(treat ~ age + educ + black + hispan + married + nodegree + 
               re74 + re75,data = lalonde,method = "nearest",caliper = 0.05)
summary(f1)
matchdata <- match.data(f)
matchdata

install.packages("writexl")
library(writexl)
matchdata$id <- 1:nrow(matchdata)
write_xlsx(matchdata,path = "D:/Downloadings__Transit/matchdata.xlsx")

#用nonrandom包做PSM
install.packages("devtools") 
devtools::install_github("cran/nonrandom")
#nonrandom包需要从github上下载
install.packages("nonrandom")
library(nonrandom)

data(stu1)
View(stu1)
stu1.ps <- pscore(data=stu1,formula = therapie ~ tgr + age)
stu1.match <- ps.match(object = stu1.ps,ratio = 2,caliper = 0.05,givenTmatchingC = FALSE,
                       matched.by = "pscore",setseed = 38902,combine.output = TRUE)
matchdata <- stu1.match$data.matched
matchdata

plot.pscore(x=stu1.ps,main = "PS disturibtion",xlab = "",par.1 = list(col="red"),
            par.0 = list(lwd=2),par.dens = list(kernel = "gaussian"))