Question

我想将非NA值保存在一行中，然后选择第一个元素列表作为新列（ncol）的最终值。

这是我想要获得的一个例子以及我无法获得的代码跑...

ID <- c(1,2,3,4) 
A <- c("A", "NA", "C", "R")
B <- c("G", "V", "NA", "T")
C <- c("NA", "NA", "NA", "Y")
D <- c("U", "W", "NA", "NA")


mydf <- data.frame(ID, A, B, C, D, ncol)


 ID  A  B  C  D ncol
1  1  A  G NA  U    A
2  2 NA  V NA  W    V
3  3  C NA NA NA    C
4  4  R  T  Y NA    R


mycol <- c(mydf$A, mydf$B, mydf$C, mydf$D)
for (i in seq(1:nrow(mydf))){
   listcolincldata <- lapply(mycol[i],[!is.na(mycols[i])])
   print(listcolincldata)
   mydf$newcol[i] <- (as.character(listcolincldata[1]))
}

Answer 1

克里斯，以下是两个示例解决方案。一个修改您的数据框以用真正的NA值替换文本“NA”，第二个使用“NA”的文本识别。两者都使用跨数据框行的apply（）（通过将margin参数设置为1）。

A <- c("A", "NA", "C", "R")
B <- c("G", "V", "NA", "T")
C <- c("NA", "NA", "NA", "Y")
D <- c("U", "W", "NA", "NA")

# First solution: convert text NA to true NA
eg1 <- data.frame(A, B, C, D, stringsAsFactors = TRUE)
eg1[eg1 == "NA"] <- NA
eg1$solution <- apply(eg1, 1, function(rw) rw[!is.na(rw)][1])
eg1

# Second solution: string recognition
eg2 <- data.frame(A, B, C, D, stringsAsFactors = TRUE)
eg2$solution <- apply(eg2, 1, function(rw) rw[rw != "NA"][1])
eg2

Answer 2

您需要使用apply来循环显示：

数据

ID <- c(1,2,3,4) 
A <- c("A", NA, "C", "R")
B <- c("G", "V", NA, "T")
C <- c(NA, NA, NA, "Y")
D <- c("U", "W", NA, NA)

mydf <- data.frame(ID, A, B, C, D)

解决方案：

#using apply is the same as lapply, but applied row wise
#na.omit will remove the NAs for each row
#then I use [2] to pick the first value after the ID column
#result is as shown in your output
mydf$ncol <- apply(mydf, 1, function(x) na.omit(x)[2])

输出：

> mydf
  ID    A    B    C    D ncol
1  1    A    G <NA>    U    A
2  2 <NA>    V <NA>    W    V
3  3    C <NA> <NA> <NA>    C
4  4    R    T    Y <NA>    R

Answer 3

另一种选择是将max.col与ties.method='first'一起使用。这将获得第一个非NA值的列索引，cbind它与行索引并获取值。

mydf$ncol <- mydf[cbind(1:nrow(mydf),max.col(!is.na(mydf[,-1]), 'first')+1L)]
mydf
#   ID    A    B    C    D ncol
#1  1    A    G <NA>    U    A
#2  2 <NA>    V <NA>    W    V
#3  3    C <NA> <NA> <NA>    C
#4  4    R    T    Y <NA>    R

数据

ID <- c(1,2,3,4) 
A <- c("A", NA, "C", "R")
B <- c("G", "V", NA, "T")
C <- c(NA, NA, NA, "Y")
D <- c("U", "W", NA, NA)

mydf <- data.frame(ID, A, B, C, D)

注意：我创建了NA作为真正的NA而不是字符串。

比较4列值以创建新列

3 个答案:

数据