我遇到了解如何使用R令人满意地重新排列数据的问题。我有数据：
id animal colour value1 value2 value3 etc.
1 pig blue 3.1 4.0 1.2
1 pig orange 3.05 4.1 NA
1 pig pink 3.05 3.9 1.25
1 pig mauve 2.95 4.0 NA
1 elephant blue 15.03 4.03 11.0
1 elephant pink 15.13 4.0 14.95
2 pig blue 4.1 ...
我希望获得类似的内容（列名称/顺序无关紧要）：
id pig_blue_value1 pig_blue_value2 pig_blue_value3 pig_orange_value1 pig_orange_value2 etc.
1 3.1 4.0 1.2 3.05 4.1
2 4.1 ...
我已经玩过reshape2的熔化和铸造，但到目前为止还没有达到任何令人满意的效果。感谢。
bluemoo <- data.frame(id=c(1,1,1,1,1,1,2),animal=c("pig","pig","pig","pig","elephant","elephant","pig"),colour=c("blue","orange","pink","mauve","blue","pink","blue"),value1=c(3.1,3.05,3.05,2.95,15.03,15.13,4.1),value2=c(4.0,4.1,3.9,4.0,4.03,4.0,NA),value3=c(1.2,NA,1.25,NA,11.0,14.95,NA))
答案 0 :(得分：4)
使用CRAN上最新的data.table版本
library(data.table) # v 1.9.6+
dcast(setDT(bluemoo), id ~ animal + colour, value.var = names(bluemoo)[-(1:3)])
# id value1_elephant_blue value1_elephant_pink value1_pig_blue value1_pig_mauve value1_pig_orange value1_pig_pink value2_elephant_blue
# 1: 1 15.03 15.13 3.1 2.95 3.05 3.05 4.03
# 2: 2 NA NA 4.1 NA NA NA NA
# value2_elephant_pink value2_pig_blue value2_pig_mauve value2_pig_orange value2_pig_pink value3_elephant_blue value3_elephant_pink
# 1: 4 4 4 4.1 3.9 11 14.95
# 2: NA NA NA NA NA NA NA
# value3_pig_blue value3_pig_mauve value3_pig_orange value3_pig_pink
# 1: 1.2 NA NA 1.25
# 2: NA NA NA NA