查找具有平局的多个列的最大值并返回列名或平局 destruct 者NA值

我正在努力寻找一个max()类型的函数，它不会返回"first '、"last"或随机，而且看起来很有希望的.max.simple()在CRAN上不再可用.

尽管多次try ，Chat GPT仍难以正确返回该内容，因此有人能帮忙吗？

library(dplyr)

# Function to find column name(s) with max value, handling tie
column_with_max <- function(...) {
  values <- c(...)
  max_value <- max(values)
  max_indices <- which(values == max_value)
  if (length(max_indices) > 1) {
    return(paste(names(values)[max_indices], collapse = ", "))
  } else {
    return(names(values)[values == max_value])
  }
}


# Sample data
data <- data.frame(group = c("A", "A", "B", "B", "C", "C"),
                   value1 = c(0, 0, 1, 0, 1, 1),
                   value2 = c(1, 1, 1, 1, 1, 1),
                   value3 = c(1, 0, 2, 0, 1, 1))

# Grouped column with max value and tie handling
result <- data %>%
  group_by(group) %>%
  summarise(max_column = column_with_max(value1, value2, value3))

期望的结果是:

output <- data.frame(group = c("A", "A", "B", "B", "C", "C"),
                   value1 = c(0, 0, 1, 0, 1, 1),
                   value2 = c(1, 1, 1, 1, 1, 1),
                   value3 = c(1, 0, 2, 0, 1, 1),
                   max_column = c("tie", "value2", "value3", "value2", "tie", "tie"))

我有多个分组变量和更大的数据，因此dspirr解决方案最有帮助.谢谢你