結合變量重塑數據

Question

我有以下數據集

D<-structure(list(P = structure(c(3L, 3L, 3L, 3L, 3L, 1L, 1L, 1L, 
1L, 1L, 2L, 2L, 2L, 2L, 2L), .Label = c("A", "B", "E"), class = "factor"), 
    ht = structure(c(4L, 2L, 3L, 5L, 1L, 4L, 2L, 3L, 5L, 1L, 
    4L, 2L, 3L, 5L, 1L), .Label = c("CH", "L65", "M65", "NOC", 
    "SPAR"), class = "factor"), hol = c(1L, 1L, 1L, 2L, 1L, 1L, 
    1L, 2L, 2L, 1L, 1L, 1L, 1L, 2L, 2L), un = c(2L, 2L, 2L, 1L, 
    2L, 2L, 2L, 1L, 1L, 2L, 2L, 2L, 2L, 1L, 1L), fu = c(3L, 3L, 
    3L, 3L, 3L, 5L, 5L, 6L, 5L, 5L, 3L, 3L, 3L, 3L, 3L)), .Names = c("P", 
"ht", "hol", "un", "fu"), class = "data.frame", row.names = c(NA, 
-15L)) 

我想使用的HT每個可能的值，然後HOL，取消，復又使其變爲從

P ht hol un fu 
1 E NOC 1 2 3 
2 E L65 1 2 3 
3 E M65 1 2 3 
5 E CH 1 2 3 
6 A NOC 1 2 5 
7 A L65 1 2 5 
9 A M65 2 1 6 

... 到

R  E A 
NOC_hol 1 1 
NOC_un 2 2 
NOC_fu 3 5 
... 
M65_fu 3 6 

我在努力尋找一種簡單的方法，它是第一次我使用重塑，所以我有點失落。

Answer 1

融化，然後重塑：

library(reshape2) 

dcast(melt(D), ht + variable ~ P) 
#  ht variable A B E 
#1 CH  hol 1 2 1 
#2 CH  un 2 1 2 
#3 CH  fu 5 3 3 
#4 L65  hol 1 1 1 
#5 L65  un 2 2 2 
#6 L65  fu 5 3 3 
#7 M65  hol 2 1 1 
#8 M65  un 1 2 2 
#9 M65  fu 6 3 3 
#10 NOC  hol 1 1 1 
#11 NOC  un 2 2 2 
#12 NOC  fu 5 3 3 
#13 SPAR  hol 2 2 2 
#14 SPAR  un 1 1 1 
#15 SPAR  fu 5 3 3 

然後，您可以paste前兩列在一起，如果你喜歡。