Question

我有两个数据帧，每个数据帧每个ID有多行。我需要根据第一个数据帧的ID和日期从第二个数据帧返回最接近的日期和相关数据 - 将相关数据添加到第一个数据帧。这也必须与第二个数据帧中的NA一起使用。示例数据：

set.seed(42)
df1 <- data.frame(ID = sample(1:3, 10, rep=T), dateTarget=(strptime((paste
    (sprintf("%02d", sample(1:30,10, rep=T)), sprintf("%02d", sample(1:12,10, rep=T)), 
     (sprintf("%02d", sample(2013:2015,10, rep=T))), sep="")),"%d%m%Y")), Value=sample(15:100, 10, rep=T))
df2 <- data.frame(ID = sample(1:3, 10, rep=T), dateTarget=(strptime((paste
     (sprintf("%02d", sample(1:30,20, rep=T)), sprintf("%02d", sample(1:12,20, rep=T)), 
     (sprintf("%02d", sample(2013:2015,20, rep=T))), sep="")),"%d%m%Y")), ValueMatch=sample(15:100, 20, rep=T))

来自base的某些内容 - split以及apply家庭的混合物？

决赛桌看起来像：

  ID dateTarget Value dateMatch ValueMatch
1  3   22-02-15    52  09-03-15         94
2  1   29-12-14    18  06-12-14         88
3  3   08-12-15    98  06-07-15         48
4  2   14-01-13    52  08-04-13         77
5  2   29-07-15    97  01-08-15         68
6  3   30-05-13    91  01-04-13         85
7  1   04-11-13    70  21-02-14         35
8  2   15-06-15    98  01-08-15         68
9  3   17-11-14    68  15-12-14         95

P.S。是否有更好的方法来生成随机日期（不使用seq.Date）？

Answer 1

以下是基于基础包的解决方案：

z <- lapply(intersect(df1$ID,df2$ID),function(id) {
   d1 <- subset(df1,ID==id)
   d2 <- subset(df2,ID==id)

   d1$indices <- sapply(d1$dateTarget,function(d) which.min(abs(d2$dateTarget - d)))
   d2$indices <- 1:nrow(d2)

   merge(d1,d2,by=c('ID','indices'))
  })

z2 <- do.call(rbind,z)
z2$indices <- NULL

print(z2)

#    ID dateTarget.x Value dateTarget.y ValueMatch
# 1   3   2015-11-14    47   2015-07-06         48
# 2   3   2015-12-08    98   2015-07-06         48
# 3   3   2015-02-22    52   2015-03-09         94
# 4   3   2014-11-17    68   2014-12-15         95
# 5   3   2013-05-30    91   2013-04-01         85
# 6   1   2013-11-04    70   2014-02-21         35
# 7   1   2014-12-29    18   2014-12-06         88
# 8   2   2013-01-14    52   2013-04-08         77
# 9   2   2015-07-29    97   2015-08-01         68
# 10  2   2015-06-15    98   2015-08-01         68

Answer 2

使用data.table，简单而优雅的解决方案：

library(data.table)

setDT(df1)
setDT(df2)

setkey(df2, ID, dateTarget)[, dateMatch:=dateTarget]
df2[df1, roll='nearest']

    ID dateTarget ValueMatch  dateMatch Value
 1:  3 2015-11-14         48 2015-07-06    47
 2:  3 2015-02-22         94 2015-03-09    52
 3:  1 2014-12-29         88 2014-12-06    18
 4:  3 2015-12-08         48 2015-07-06    98
 5:  2 2013-01-14         77 2013-04-08    52
 6:  2 2015-07-29         68 2015-08-01    97
 7:  3 2013-05-30         85 2013-04-01    91
 8:  1 2013-11-04         35 2014-02-21    70
 9:  2 2015-06-15         68 2015-08-01    98
10:  3 2014-11-17         95 2014-12-15    68

Answer 3

这是我根据接受的答案使用dplyr的方法。我希望在分组列上有更多的自由。


match_by_group_date <- function(df1, df2, grp, datecol) {
  
  grp1 <- df1 %>% pull({{grp}}) %>% unique()
  grp2 <- df2 %>% pull({{grp}}) %>% unique()
  
  li <-
  lapply(intersect(grp1, grp2), function(tt) {
    d1 <- filter(df1, {{grp}}== tt)
    d2 <- filter(df2, {{grp}}==tt) %>% mutate(indices = 1:n())
    d2_date <- d2 %>% pull({{datecol}}) %>% as.POSIXct()
    print(d2_date)
    d1 <- mutate(d1, indices = map_dbl({{datecol}}, function(d) which.min(abs(d2_date - as.POSIXct(d)))))
    
    left_join(d1,d2, by=c(quo_name(enquo(grp)), "indices"))
  })
  
  # bind rows
  return(bind_rows(li))
}

Answer 4

我们也可以使用dplyr来实现此功能。

library(dplyr)

left_join(df1, df2, by = "ID") %>%
  mutate(dateDiff = abs(dateTarget.x - dateTarget.y)) %>%
  group_by(ID, dateTarget.x) %>%
  filter(dateDiff == min(dateDiff))

合并最近的日期以及来自另一个数据帧的相关变量

4 个答案: