如何准备for循环

Question

我的 adsl 有几个数字变量，我想对它们进行汇总以获得汇总数据集。下面是我的代码，我以 AGE 和 STAGEN 变量为例：

adsl <- adsl %>% group_by(ACTARM)

AGE <- summarise(adsl,
    param='AGE',
    mean=mean(AGE, na.rm = TRUE),
    sd = sd(AGE, na.rm = TRUE),
    median = quantile(AGE, probs = 0.5),
    min = min(AGE),
    max = max(AGE)
  )
STAGEN <- summarise(adsl,
                    param='STAGEN',
                    mean=mean(STAGEN, na.rm = TRUE),
                    sd = sd(STAGEN, na.rm = TRUE),
                    median = quantile(STAGEN, probs = 0.5),
                    min = min(STAGEN),
                    max = max(STAGEN)
)
result <- rbind(AGE, STAGEN)

我的问题是如何编写一个for循环来替换重复的代码

for (i in c('AGE', 'STAGEN')){
  i <- adsl %>% 
    summarise(
      param=i,
      mean=mean(adsl[i], na.rm = TRUE),
      sd = sd(adsl[i], na.rm = TRUE),
      median = quantile(adsl[i], probs = 0.5),
      min = min(adsl[i]),
      max = max(adsl[i])
    )
}

希望有人能帮助我

Answer 1

我想把我的问题说清楚。我准备了一个adsl数据集。我想获得一个摘要数据集。

当我用 AGE 检查我的函数时，它运行良好，但是如果我想使用 for 循环处理多个变量，输出中的 param 列显示“i”，实际上我希望它显示“AGE”，“HEIGHT”， '重量'

SUBJID <- c(paste0('S', 1:10))
ACTARM <- c(paste0('TRT', rep(1:2, 5)))
AGE <- c(23:32)
HEIGHT <- c(160:169)
WEIGHT <- c(60:69)
adsl <- tibble(
  SUBJID=SUBJID,
  ACTARM=ACTARM,
  AGE=AGE,
  WEIGHT=WEIGHT,
  HEIGHT=HEIGHT
)

sum_num <- function(data, var, group_var = ACTARM ){
  var_name <- tail(as.character(substitute(var)), 1)
  data %>% group_by({{group_var}}) %>% 
    summarize(
      param=var_name ,
      n = sum(!is.na({{var}})),
      mean = mean({{var}}, na.rm = TRUE),
      sd = sd({{var}}, na.rm = TRUE),
      median = quantile({{var}}, probs = 0.5, type = 3),
      min = min({{var}})
    )
}

#check the function with one varaible
chk <- sum_num(adsl, AGE)

#check the function with multiple variables
output <- tibble()
for (i in c('AGE', 'HEIGHT', "WEIGHT")){
  temp <- sum_num(adsl, .data[[i]])
  output <- rbind(output, temp)
}