你好,我需要编程R的帮助。我有data.frame B,有四列
x<- c(1,2,1,2,1,2,1,2,1,2,1,2,.......etc.)
y<-c(5,5,8,8,12,12,19,19,30,30,50,50,...etc.)
z<- c(2018-11-08,2018-11-08,2018-11-09,2018-11-09,2018-11-11,2018-11-11,2018-11-20,2018-11-20,2018-11-29,2018-11-29,2018-11-30,2018-11-30,.......etc.)
m<-c(0,1,1,0,1,1,0,1,0,1,0,1,...etc.)
200万行,我需要创建下一列。接下来的列应为
t<-c(0,1,0,0,0,0,0,1,0,1,0,1,....)
循环代码类似于
B$t[1]=ifelse(B$y[i]==B$y[i+1] & B$z[i]==B$z[i+1] & B$x[i]==2 & B$m[1]==1,1,0)
for (i in 2:length(B$z))
{
B$t[i]<-ifelse(B$y[i]==B$y[i-1] & B$z[i]==B$z[i-1] & B$x[i]==2 & B$m[i]==1 & B$m[i]!=B$m[i-1],1,0)
}
我不想使用循环循环。
我在R中使用基本软件包。
当我有data.frame E时,我有一个新问题
x<- c(1,2,3,1,2,3,1,2,3,1,2,3,.......etc.)
y<-c(5,5,5,8,8,8,12,12,12,,19,19,19,30,30,30,50,50,50,...etc.)
z<- c(2018-11-08,2018-11-08,2018-11-08,2018-11-09,2018-11-09,2018-11-09,2018-11-11,2018-11-11,2018-11-11,2018-11-20,2018-11-20,2018-11-20,2018-11-29,2018-11-29,2018-11-29,2018-11-30,2018-11-30,2018-11-30,.......etc.)
m<-c(0,1,1,0,0,1,0,1,0,1,0,1,0,0,1...etc.)
200万行,我需要创建下一列。接下来的列应为
t<-c(0,1,0,0,1,....)
循环代码类似于
E$t[1]=ifelse(E$y[i]==E$y[i+1] & E$z[i]==E$z[i+1] & E$x[1]==2 & E$m[1]==1,1,0)
E$t[2]=ifelse(E$y[i]==E$y[i+1] & E$z[i]==E$z[i+1] & E$x[2]==3 & E$m[2]==1,1,0)
for (i in 3:length(E$y))
{
E$t[i]<-ifelse(E$y[i]==E$y[i-2] & E$z[i]==E$z[i-2] & E$x[i]==3 & E$m[i]==1 &
E$m[i-1]==0 & E$m[i-2]==0,1,0)
}
我不想使用循环循环。
我在R中使用基本软件包。
答案 0 :(得分:1)
通过dplyr
,您可以使用if_else
和lag
:
library(dplyr)
dat %>%
mutate(t = if_else(
y == lag(y) & z == lag(z) & x == 2 & m == 1 & m != lag(m), 1, 0)
) # mutate lets you create a new variable in dat (named t here)
# x y z m t
# 1 1 5 2018-11-08 0 0
# 2 2 5 2018-11-08 1 1
# 3 1 8 2018-11-09 1 0
# 4 2 8 2018-11-09 0 0
# 5 1 12 2018-11-11 1 0
# 6 2 12 2018-11-11 1 0
# 7 1 19 2018-11-20 0 0
# 8 2 19 2018-11-20 1 1
# 9 1 30 2018-11-29 0 0
# 10 2 30 2018-11-29 1 1
# 11 1 50 2018-11-30 0 0
# 12 2 50 2018-11-30 1 1
数据:
x<- c(1,2,1,2,1,2,1,2,1,2,1,2)
y<-c(5,5,8,8,12,12,19,19,30,30,50,50)
z<- c("2018-11-08","2018-11-08","2018-11-09","2018-11-09","2018-11-11","2018-11-11","2018-11-20","2018-11-20","2018-11-29","2018-11-29","2018-11-30","2018-11-30")
m<-c(0,1,1,0,1,1,0,1,0,1,0,1)
dat <- data.frame(x, y, z, m)
答案 1 :(得分:1)
以下是基于R
的解决方案:
N <- nrow(B)
B$t <- ifelse(B$y==c(NA, B$y[-N]) & B$z==c(NA, B$z[-N]) & B$x==2 & B$m==1 & B$m!=c(NA, B$m[-N]), 1, 0)
这是data.table
的解决方案:
library("data.table")
B <- data.table(
x= c(1,2,1,2,1,2,1,2,1,2,1,2), y= c(5,5,8,8,12,12,19,19,30,30,50,50),
z= c("2018-11-08", "2018-11-08", "2018-11-09", "2018-11-09", "2018-11-11", "2018-11-11", "2018-11-20",
"2018-11-20", "2018-11-29", "2018-11-29", "2018-11-30", "2018-11-30"),
m= c(0,1,1,0,1,1,0,1,0,1,0,1)
)
B[, t := ifelse(y==c(NA, y[- .N]) & z==c(NA, z[- .N]) & x==2 & m==1 & m!=c(NA, m[- .N]), 1, 0)]
或(如果可以接受逻辑)
B[, t := (y==c(NA, y[- .N]) & z==c(NA, z[- .N]) & x==2 & m==1 & m!=c(NA, m[- .N]))]
或使用shift()
B[, t := (y==shift(y) & z==shift(z) & x==2 & m==1 & m!=shift(m))]