R 基于先前出现的增加变量

Question

我有一个按日期排序的餐厅检查数据框。对于每个观察，我想添加两个额外的变量来记录这家餐厅总共进行了多少次检查，以及他们失败了多少次。我想避免使用 for 循环，但我不知道该怎么做。本质上，我目前有一个数据框，由下面数据框的前三列组成，我想添加最后两列。

初始数据帧

    Restaurant_ID    Date         Result
    1                01/02/2011   Pass 
    2                02/05/2011   Pass
    3                04/07/2011   Fail
    1                09/05/2011   Fail
    2                03/13/2012   Pass
    1                08/25/2012   Fail
    2                09/25/2012   Pass
    3                01/05/2013   Pass

所需的输出 1

Restaurant_ID    Date         Result   total_inspect  failed_inspect
1                01/02/2011   Pass     1              0
2                02/05/2011   Pass     1              0
3                04/07/2011   Fail     1              1
1                09/05/2011   Fail     2              1
2                03/13/2012   Pass     2              0
1                08/25/2012   Fail     3              2
2                09/25/2012   Pass     3              0
3                01/05/2013   Pass     2              1

编辑：我意识到我实际上希望最后两列反映当前观察之前的总检查次数和失败检查次数。所以我真正想要的是

所需输出 2

    Restaurant_ID    Date         Result   past_inspect  past_failed_inspect
    1                01/02/2011   Pass     0              0
    2                02/05/2011   Pass     0              0
    3                04/07/2011   Fail     0              0
    1                09/05/2011   Fail     1              0
    2                03/13/2012   Pass     1              0
    1                08/25/2012   Fail     2              1
    2                09/25/2012   Pass     2              0
    3                01/05/2013   Pass     1              1

Answer 1

此解决方案使用包 tidyverse 和 lubridate 中的函数。

# Create the example data frame
dt1 <- read.csv(text = "Restaurant_ID,Date,Result
1,01/02/2011,Pass
2,02/05/2011,Pass
3,04/07/2011,Fail
1,09/05/2011,Fail
2,03/13/2012,Pass
1,08/25/2012,Fail
2,09/25/2012,Pass
               3,01/05/2013,Pass",
               stringsAsFactors = FALSE)

# Load packages
library(tidyverse)
library(lubridate)

dt2 <- dt1 %>%
  # Convert the Date column to Date class
  mutate(Date = mdy(Date)) %>%
  # Sort the data frame based on Restaurant_ID and Date
  arrange(Restaurant_ID, Date) %>%
  # group the data by each restaurant ID
  group_by(Restaurant_ID) %>%
  # Create a column showing total_inspect
  mutate(total_inspect = 1:n()) %>%
  # Create a column showing fail_result, fail is 1, pass is 0
  mutate(fail_result = ifelse(Result == "Fail", 1, 0)) %>%
  # Calculate the cumulative sum of fail_result
  mutate(failed_inspect = cumsum(fail_result)) %>%
  # Remove fail_result
  select(-fail_result) %>%
  # Sort the data frame by Date
  arrange(Date)

编辑：计算过去的检查和失败次数

dt3 <- dt2 %>%
  mutate(past_inspect = ifelse(total_inspect == 0, 0, total_inspect - 1)) %>%
  mutate(past_failed_inspect = ifelse(Result == "Fail" & failed_inspect != 0, 
                                      failed_inspect - 1,
                                      failed_inspect)) %>%
  select(-total_inspect, -failed_inspect)