【R语言】年龄性别频数匹配 挑选样本 病例对照研究，对年龄性别进行频数匹配

病例对照研究，对年龄性别进行频数匹配

示例数据

示例数据下载

ID	group	sex	age
A1	Control	女	90
A10	Control	男	89
A100	Control	男	85
A1000	Control	男	74
A1001	Case	女	74
A1002	Case	男	74
A1003	Case	女	74
A1004	Case	男	74
A1005	Control	男	74

data <- read.csv("年龄性别频数匹配.csv")

## 可以先对年龄进行分层处理
data$agegroup <- cut(data$age,c(0,10,20,30,40,50,60,70,80,90,Inf),
                     right =FALSE # 选取区间开口[ ]情况
                     )
# 生成list数据，按每一个list[?]为同样sex和age
dataSplit <- split(data, list(data$sex, data$age))

# 单个list提取
A <- as.data.frame(data_case[[1]])
data1 <- A[A$group == "Case", ]
data2 <- A[A$group == "Control", ]
num <- as.numeric(table(A$group == "Case")[2])
set.seed(123) # 设置随机数目，多次重复保持一致
data3 <- data2[sample(1:NROW(data2), num, replace = FALSE), ]
assign(paste0("dataFinal", 1), rbind(data1, data3))

# 循环提取
for (i in 1:length(dataSplit)) {
  A <- as.data.frame(data_case[[i]])
    data1 <- A[A$group == "Case", ]
    data2 <- A[A$group == "Control", ]
    numCase <- nrow(data1)
    numControl <- nrow(data2)
    if (numCase > 0 & numControl>0 & numControl-numCase>=0) {
      na.omit <- TRUE
      set.seed(123) # 设置随机数目，多次重复保持一致 【可以沉默】
      data3 <- data2[sample(1:numControl, numCase, replace = FALSE), ]
      assign(paste0("dataFinal", i), rbind(data1, data3))
    }
    if (numCase > 0 & numControl>0 & numControl-numCase<0) {
      na.omit <- TRUE
      set.seed(123) # 设置随机数目，多次重复保持一致 【可以沉默】
      data3 <- data1[sample(1:numCase, numControl, replace = FALSE), ]
      assign(paste0("dataFinal", i), rbind(data2, data3))
    } 
}

# 多个数据框合并
multimerge<-function(dat=list(),...){
  if(length(dat)<2)return(as.data.frame(dat))
  mergedat<-dat[[1]]
  dat[[1]]<-NULL
  for(i in dat){
    mergedat<-merge(all=TRUE,mergedat,i,...)
  }
  return(mergedat)
}

files=ls(pattern = "dataFinal")
listALL=list()
# 数据框合成list
for (i in 1:length(ls(pattern = "dataFinal"))) {
  eval(parse(text = paste0("listALL","[[",i,"]]", " <- ",files[i])))
}

dataALL <- multimerge(listALL)
# 排序
dataALL=dplyr::arrange(dataALL,age,sex,group)
# 导出
write.csv(dataALL,"dataALL.csv",row.names = F)