R语言宽数据框子集化：基于指定数据集生成3个子表需求

没问题，我来帮你处理这个数据集！先确认你的示例数据结构：

df <- structure(list(
  Variable = c("A", "A", "A", "B", "B", "B", "C", "C", "C"),
  Day1 = c("1", "1", "-1", "1", "1", "-1", "1", "1", "1"),
  Day2 = c("1", "1", "-1", "1", "1", "-1", "1", "1", "1"),
  Day3 = c("1", "1", "1", "1", "1", "-1", "1", "1", "1"),
  Day4 = c("1", "1", "1", "1", "1", "1", "1", "1", "-1"),
  Day5 = c("1", "1", "1", "1", "1", "1", "1", "1", "-1")
), class = "data.frame", row.names = c(NA, -9L))

接下来我会用dplyr和tidyr包生成三个实用的统计表格，如果你还没安装这两个包，先运行install.packages(c("dplyr", "tidyr"))即可。

表格1：按Variable分组的每日1/-1出现次数 #

这个表格清晰展示每个变量在每一天中，值为1和-1的样本数量：

library(dplyr)
library(tidyr)

table1 <- df %>%
  pivot_longer(cols = starts_with("Day"), names_to = "Day", values_to = "Value") %>%
  group_by(Variable, Day, Value) %>%
  summarise(Count = n(), .groups = "drop") %>%
  pivot_wider(names_from = Value, values_from = Count, values_fill = 0)

print(table1)

输出表格：

表格2：每个Variable首次出现-1的日期 #

这个表格帮你快速定位每个变量第一次出现-1的时间，未出现过-1的标记为"无"：

table2 <- df %>%
  pivot_longer(cols = starts_with("Day"), names_to = "Day", values_to = "Value") %>%
  filter(Value == "-1") %>%
  group_by(Variable) %>%
  summarise(First_Negative_Day = min(Day), .groups = "drop") %>%
  right_join(unique(df %>% select(Variable)), by = "Variable") %>%
  mutate(First_Negative_Day = ifelse(is.na(First_Negative_Day), "无", First_Negative_Day))

print(table2)

输出表格：

表格3：每日各Variable的1值占比 #

这个表格展示每个变量在每一天中，值为1的样本占该变量当日总样本的百分比（保留1位小数）：

table3 <- df %>%
  pivot_longer(cols = starts_with("Day"), names_to = "Day", values_to = "Value") %>%
  group_by(Variable, Day) %>%
  summarise(Positive_Ratio = mean(Value == "1") * 100, .groups = "drop") %>%
  mutate(Positive_Ratio = round(Positive_Ratio, 1))

print(table3)

输出表格：

内容的提问来源于stack exchange，提问作者flightless13wings