R：如何根据序列和其他列中的值获取数据帧列中的值

我有一个数据帧，df。我按v1和v2排序的数据帧。 对于v1中的每组唯一值（示例数据中的值1、2和3），我想计算一个新变量v5

v5的值取决于v3和v4的值： 如果v3==“新建”，则v5==v4。 如果v3==“Old”，则v5获取行中v4的值，其中v3中的前一个值等于“New”。都在v1的同一“组”内

样本数据：

df <- data.frame(v1=c(1,1,1,2,2,2,3,3,3,3), 
             v2=c(1,2,3,1,2,3,1,2,3,4),
             v3=c("New", "Old", "Old","New", "Old", "New","New", "New", "Old","Old"),
             v4=c("A","B","C","X","Y","Z","A","B","C","D")) 


v1 v2  v3 v4
1  1 New  A  
1  2 Old  B
1  3 Old  C
2  1 New  X
2  2 Old  Y
2  3 New  Z
3  1 New  A
3  2 New  B
3  3 Old  C
3  4 Old  D

---

我们可以尝试使用

data.table

。将'data.frame'转换为'data.table'（

setDT（df）

），按'v1'分组，我们用NA替换'v3'中与'Old'值相对应的'v4'元素，然后使用

NA.locf

（来自

库（动物园）

）将

NA

值替换为前面的非NA值，分配（

：=/code>）用于创建新列“v5”的输出

library(data.table)
library(zoo)
setDT(df)[, v5:= na.locf(replace(v4, v3=='Old', NA)) , by = v1]
df
#    v1 v2  v3 v4 v5
# 1:  1  1 New  A  A
# 2:  1  2 Old  B  A
# 3:  1  3 Old  C  A
# 4:  2  1 New  X  X
# 5:  2  2 Old  Y  X
# 6:  2  3 New  Z  Z
# 7:  3  1 New  A  A
# 8:  3  2 New  B  B
# 9:  3  3 Old  C  B
#10:  3  4 Old  D  B


或者我们可以使用
ave
from
base R

df$v5 <- with(df, ave(replace(v4, v3=='Old', NA),v1, FUN= na.locf)) 

df$v5也可以使用
dplyr
包

library(dplyr)
library(zoo)
df <- data.frame(v1=c(1,1,1,2,2,2,3,3,3,3), 
                 v2=c(1,2,3,1,2,3,1,2,3,4),
                 v3=c("New", "Old", "Old","New", "Old", "New","New", "New", "Old","Old"),
                 v4=c("A","B","C","X","Y","Z","A","B","C","D"),
                 stringsAsFactors = FALSE) 
df %>% 
  group_by(v1) %>%
  mutate(v5=ifelse(v3=="New", v4, NA),
         v5=na.locf(v5))
# Source: local data frame [10 x 5]
# Groups: v1 [3]
# 
#       v1    v2    v3    v4    v5
#    (dbl) (dbl) (chr) (chr) (chr)
# 1      1     1   New     A     A
# 2      1     2   Old     B     A
# 3      1     3   Old     C     A
# 4      2     1   New     X     X
# 5      2     2   Old     Y     X
# 6      2     3   New     Z     Z
# 7      3     1   New     A     A
# 8      3     2   New     B     B
# 9      3     3   Old     C     B
# 10     3     4   Old     D     B

库（dplyr）
图书馆（动物园）
df%
分组依据（v1）%>%
变异（v5=ifelse（v3=New），v4，NA），
v5=na.locf（v5））
#来源：本地数据帧[10 x 5]
#分组：v1[3]
# 
#v1 v2 v3 v4 v5
#（dbl）（dbl）（chr）（chr）（chr）
#1新的A
#2 1 2老B A
#3 1 3旧C A
#4 2 1新X X
#5 2 2老Y X
#6 2 3新Z Z
#7 3 1新A
#8 3 2新B
#9 3旧C B
#10 3 4老D B
太好了。Thx@docendo discimus
library(dplyr)
library(zoo)
df <- data.frame(v1=c(1,1,1,2,2,2,3,3,3,3), 
                 v2=c(1,2,3,1,2,3,1,2,3,4),
                 v3=c("New", "Old", "Old","New", "Old", "New","New", "New", "Old","Old"),
                 v4=c("A","B","C","X","Y","Z","A","B","C","D"),
                 stringsAsFactors = FALSE) 
df %>% 
  group_by(v1) %>%
  mutate(v5=ifelse(v3=="New", v4, NA),
         v5=na.locf(v5))
# Source: local data frame [10 x 5]
# Groups: v1 [3]
# 
#       v1    v2    v3    v4    v5
#    (dbl) (dbl) (chr) (chr) (chr)
# 1      1     1   New     A     A
# 2      1     2   Old     B     A
# 3      1     3   Old     C     A
# 4      2     1   New     X     X
# 5      2     2   Old     Y     X
# 6      2     3   New     Z     Z
# 7      3     1   New     A     A
# 8      3     2   New     B     B
# 9      3     3   Old     C     B
# 10     3     4   Old     D     B