使用 mapply 为切点添加标签

Adding labels to cut points with mapply

我有数据如下:

library(data.table)
library(Hmisc)
dat <- structure(list(Inc= c(120995, 238097.2, 103993.9, 93801, 
255422.769863014, 257038.28739726, 1045388.66465753, 1040685.40328767, 
715660.096547945, 484324), value = list(c(0, 15659.7, 78212.8, 
419000, 1e+09), c(0, 16136.4, 89658.3, 464800, 1e+09), c(25075, 
98208.05, 164627, 276586.59, 1e+09), c(25003.35, 91842, 149675.7, 
256661.25, 1e+09), c(25024.7, 54664.75, 123463.2, 258610.05, 
1e+09), c(25081, 56070, 120899.15, 282341.03, 1e+09), c(44852.9, 
299214, 514994.25, 802947.38, 1e+09), c(36074, 311042.59, 528348.81, 
754085.44, 1e+09), c(25540.8, 181958.59, 326804.69, 538702.19, 
1e+09), c(28311.2, 175799.41, 316315, 507802.56, 1e+09))), row.names = c(NA, 
-10L), class = c("data.table", "data.frame"))

value 列中的每一行都有 5 个值。使用 cut2,我可以将这些值用作切点,如下所示:

dat <- setDT(dat)[, cats:= mapply(Hmisc:: cut2, Inc, value, oneval=FALSE)]

问题是 ?cut2 似乎不允许使用标签。结果我尝试了 cut,但无论我做什么,它都说标签和类别不匹配。

dat <- setDT(dat)[, cats := mapply(cut, Inc, breaks=value, labels=c("cat1", "cat2", "cat3", "cat4"))]

期望的输出:

          Inc                                                            value cats
 1:  120995.0      0.00000e+00,1.56597e+04,7.82128e+04,4.19000e+05,1.00000e+09 cat3
 2:  238097.2      0.00000e+00,1.61364e+04,8.96583e+04,4.64800e+05,1.00000e+09 cat3
 3:  103993.9 2.507500e+04,9.820805e+04,1.646270e+05,2.765866e+05,1.000000e+09 cat2
 ...
10:  484324.0      28311.2,    175799.4,    316315.0,    507802.6,1000000000.0 cat3

因为您已经在 value 列中有了间隔(可能是其他地方 cut() 的结果),您可以按如下方式使用 findInterval()

setDT(dat)[, cats:= paste0('cat', mapply(FUN = findInterval, x = Inc, vec = value))]

(确保 value 中的分隔符是有序的)

你可以使用 santoku::chop():

setDT(dat)[, cats := mapply(chop, Inc, value, 
                       MoreArgs = list(labels = letters[1:4]))]