如何选择每周数据并执行进一步的计算，然后将它们分组为r

Question

我有以下数据：

 Date          Price
 2-Jul-13      20
 3-Jul-13      22
 4-Jul-13      30
 5-Jul-13      18     
 8-Jul-13      12
 9-Jul-13      24
10-Jul-13      28
11-Jul-13      14

产出必须是：

  Date        Price   day_diff   week_diff
 2-Jul-13      20         0          4  
 3-Jul-13      22         2
 4-Jul-13      30         8
 5-Jul-13      18       -12
 8-Jul-13      12        -6
 9-Jul-13      24        12         -4  
10-Jul-13      28         4
11-Jul-13      14        -14 
12-Jul-13      18         4
15-Jul-13      12        -6
16-Jul-13      20         8         ....  
    ....      

要计算day_diff，第一个值为0，然后为22-20 =2，依此类推。为了计算week_diff，下一周从9-7月13日开始，所以24-20 =4，同样，下周从16-7月-13，所以20-24 = -4开始，等等.

请帮我处理这个。

Answer 1

请提供您的数据与dput()在未来。

数据：

read.table(
  text = " Date        Price
2-Jul-13      20
3-Jul-13      22
4-Jul-13      30
5-Jul-13      18
8-Jul-13      12
9-Jul-13      24
10-Jul-13      28
11-Jul-13      14
12-Jul-13      18
15-Jul-13      12
16-Jul-13      20 ",
  header = T
) -> df

解决方案：

library(tidyverse)
library(lubridate)

df %>%
  mutate(
    Date = dmy(Date),
    day_diff = Price - lag(Price),
    week_date = floor_date(Date, unit = 'week', week_start = 2),
    week_number = ifelse(Date == week_date, Price, 0),
    week_diff = lead(week_number, 5) - week_number
  ) %>%
  select(Date, Price, day_diff, week_diff)-> output_df

输出

> output_df
         Date Price day_diff week_diff
1  2013-07-02    20       NA         4
2  2013-07-03    22        2         0
3  2013-07-04    30        8         0
4  2013-07-05    18      -12         0
5  2013-07-08    12       -6         0
6  2013-07-09    24       12        -4
7  2013-07-10    28        4        NA
8  2013-07-11    14      -14        NA
9  2013-07-12    18        4        NA
10 2013-07-15    12       -6        NA
11 2013-07-16    20        8        NA