我有4列的数据框。
我想按列age_group将数据帧拆分为单独的数据帧，并计算从segment_2018到segment_2020的过渡。因此，结果应该是由age_group产生的几个数据表（取决于table(df$segment_2018, df$segment_2020)值的数量）。有什么想法吗？
数据样本：
structure(list(cust_id = c(5689748L, 1256987L, 8596263L, 4152659L,
4589521L, 0125698L, 2896359L, 2045975L, 3759826L, 4625831L, 1875964L,
6132852L, 8365472L, 1287465L, 9765287L, 9357452L, 8725691L, 4051697L,
5783105L, 6040870L), segment_2018 = c("256", "258", "259", "2061",
"2061", "2061", "7", "256", "259", "1029", "256", "258", "256",
"67", "12", "258", "4115", "4115", "13", "1029"), age_group = c("58_59",
"70_71", "62_63", "56_57", "62_63", "0", "46_47", "52_53", "52_53",
"52_53", "56_57", "50_51", "0", "52_53", "50_51", "62_63", "62_63",
"70_71", "44_45", "50_51"), segment_2020 = c("256", "258", "256",
"2061", "17", "0", "7", "17", "133", "528", "256", "258", "0",
"67", "12", "258", "133", "4114", "12", "1029")), row.names = c(NA,
20L), class = "data.frame")
cust_id segment_2018 age_group segment_2020
1 5689748 256 58_59 256
2 1256987 258 70_71 258
3 8596263 259 62_63 256
4 4152659 2061 56_57 2061
5 4589521 2061 62_63 17
6 125698 2061 0 0
7 2896359 7 46_47 7
8 2045975 256 52_53 17
9 3759826 259 52_53 133
10 4625831 1029 52_53 528
11 1875964 256 56_57 256
12 6132852 258 50_51 258
13 8365472 256 0 0
14 1287465 67 52_53 67
15 9765287 12 50_51 12
16 9357452 258 62_63 258
17 8725691 4115 62_63 133
18 4051697 4115 70_71 4114
19 5783105 13 44_45 12
20 6040870 1029 50_51 1029
预期输出：
structure(c(1859L, 3661L, 214L, 106L, 107L, 209L, 341L, 1770L,
16343L, 106881L, 5078L, 317L, 593L, 8237L, 1106L, 271L, 402L,
285L, 422L, 428L, 115L, 365L, 40507L, 11700L, 132L, 50L, 815L,
375L, 189L, 998L, 14207L, 3171L, 882L, 307L, 948L, 7774L, 1985L,
1414L, 2025L, 750L, 929L, 947L, 21L, 810L, 905L, 14358L, 4L,
0L, 97L, 115L, 21L, 547L, 12926L, 2285L, 154L, 24L, 1120L, 1851L,
346L, 215L, 122L, 79L, 98L, 310L, 1L, 72L, 502L, 251L, 10264L,
1837L, 85L, 33L, 14L, 17L, 240L, 185L, 74L, 21L, 48L, 401L, 225L,
111L, 115L, 23L, 57L, 77L, 94L, 187L, 313L, 150L, 206L, 5228L,
78L, 35L, 13L, 2L, 143L, 120L, 66L, 18L, 23L, 269L, 136L, 64L,
106L, 19L, 48L, 66L, 1057L, 121L, 1531L, 563L, 51L, 33L, 2922L,
266L, 86L, 24L, 305L, 74L, 513L, 311L, 85L, 875L, 1068L, 291L,
315L, 48L, 1116L, 902L, 15L, 197L, 497L, 418L, 66L, 28L, 439L,
1517L, 35L, 26L, 491L, 233L, 170L, 92L, 238L, 597L, 325L, 122L,
339L, 117L, 120L, 1209L, 32L, 91L, 236L, 739L, 4L, 0L, 43L, 26L,
5345L, 1443L, 182L, 171L, 432L, 190L, 69L, 823L, 202L, 7L, 138L,
72L, 23L, 72L, 0L, 15L, 44L, 274L, 3L, 1L, 3L, 4L, 68L, 4170L,
141L, 575L, 185L, 31L, 30L, 122L, 1L, 5L, 4L, 2L, 4L, 8L, 0L,
11L, 1891L, 6236L, 75L, 31L, 126L, 192L, 12L, 429L, 44940L, 11113L,
544L, 93L, 704L, 4536L, 414L, 529L, 175L, 88L, 266L, 385L, 26L,
476L, 1882L, 2654L, 84L, 48L, 78L, 186L, 171L, 1112L, 15439L,
64342L, 1394L, 174L, 531L, 5187L, 608L, 178L, 313L, 193L, 256L,
383L, 22L, 211L, 182L, 83L, 44L, 18L, 215L, 78L, 51L, 70L, 139L,
117L, 16367L, 912L, 85L, 182L, 71L, 104L, 327L, 99L, 214L, 233L,
15L, 142L, 136L, 49L, 16L, 10L, 194L, 63L, 65L, 49L, 63L, 35L,
2214L, 3989L, 35L, 124L, 38L, 6L, 166L, 39L, 43L, 128L, 13L,
49L, 159L, 2751L, 1L, 2L, 27L, 63L, 1L, 37L, 1371L, 444L, 85L,
13L, 1098L, 308L, 123L, 52L, 84L, 60L, 27L, 270L, 0L, 17L, 3610L,
10976L, 80L, 32L, 417L, 383L, 915L, 2046L, 29728L, 7587L, 1804L,
468L, 818L, 72508L, 7699L, 729L, 1357L, 735L, 669L, 960L, 17L,
448L, 1746L, 9166L, 38L, 13L, 526L, 232L, 250L, 212L, 4648L,
1099L, 433L, 129L, 859L, 16061L, 9197L, 471L, 1658L, 594L, 431L,
722L, 10L, 241L, 1062L, 864L, 87L, 4L, 177L, 61L, 2L, 7L, 473L,
177L, 105L, 2L, 129L, 810L, 487L, 3680L, 253L, 92L, 338L, 183L,
6L, 417L, 4791L, 3960L, 44L, 28L, 240L, 279L, 304L, 99L, 1559L,
545L, 947L, 332L, 1396L, 4115L, 4226L, 533L, 3921L, 624L, 222L,
1234L, 14L, 235L, 763L, 1480L, 5L, 1L, 46L, 84L, 123L, 41L, 628L,
165L, 124L, 46L, 601L, 1012L, 813L, 102L, 253L, 561L, 51L, 320L,
1L, 44L, 591L, 227L, 17L, 5L, 584L, 74L, 15L, 7L, 241L, 84L,
163L, 47L, 18L, 497L, 288L, 305L, 44L, 15L, 3920L, 146L, 5L,
109L, 1613L, 1577L, 61L, 32L, 1657L, 883L, 108L, 44L, 1195L,
465L, 493L, 219L, 951L, 1555L, 1275L, 296L, 1704L, 460L, 368L,
2584L, 25L, 199L, 254L, 67L, 232L, 276L, 176L, 82L, 6L, 5L, 95L,
110L, 73L, 32L, 30L, 170L, 126L, 54L, 98L, 13L, 53L, 144L, 4957L,
147L, 354L, 198L, 2424L, 98L, 53L, 26L, 6L, 14L, 168L, 133L,
53L, 6L, 37L, 323L, 127L, 427L, 81L, 25L, 44L, 51L, 37L, 12899L
), .Dim = 22:23, .Dimnames = structure(list(c("1029", "1031",
"12", "13", "133", "17", "2056", "2060", "2061", "256", "258",
"259", "265", "4114", "4115", "5", "528", "529", "65", "67",
"7", "9"), c("0", "1029", "1031", "12", "13", "133", "17", "2056",
"2060", "2061", "256", "258", "259", "265", "4114", "4115", "5",
"528", "529", "65", "67", "7", "9")), .Names = c("", "")), class = "table")
我们可以分割数据并在每个组上使用table。这可以通过split和lapply完成。
temp <- lapply(split(df, df$age_group), function(x)
table(x$segment_2018, x$segment_2020))
或使用by：
temp <- by(df, df$age_group, function(x) table(x$segment_2018, x$segment_2020))
这将返回表列表。通常，最好将它们保留为列表，因为它更易于管理并且不会使全局环境混乱，但是如果要将它们作为单独的对象，我们可以使用list2env。
#As temp have names with numbers prefixing "table" to it.
names(temp) <- paste0('table_', names(temp))
list2env(temp, .GlobalEnv)
我们也可以这样做
temp <- lapply(split(df, df$age_group), function(x)
table(x[c('segment_2018', 'segment_2020')]))