我正在使用代码从较大的数据集中过滤出较小的数据集。我选择 24 个月以下的儿童和另一个变量 (b9)，该变量表示孩子是否与母亲住在一起。 然而，在某些情况下，可能有超过一个 24 个月以下的孩子与母亲一起生活，在这种情况下，我只想选择按 caseid 和出生史 (bidx) 排列的最小的孩子。我正在使用以下代码，但它给了我一个空数据框。我没有收到任何错误，代码运行但它返回一个空数据框，但它不应该返回。这是一个逻辑错误。
###SAMPLE Dataframe
structure(list(v001 = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1), v002 = c(107,
107, 107, 107, 113, 113, 113, 119, 125, 131, 137, 137, 137, 143,
143, 143, 149, 149, 15, 15, 15, 15, 15, 15, 156, 156, 21, 21,
3, 3, 33, 39, 46, 52, 52, 52, 58, 64, 64, 64, 64, 76, 76, 76,
76, 82, 82, 82, 82, 82), caseid = c(" 1 107 2", " 1 107 2",
" 1 107 10", " 1 107 10", " 1 113 2", " 1 113 2",
" 1 113 2", " 1 119 3", " 1 125 2", " 1 131 2",
" 1 137 2", " 1 137 2", " 1 137 2", " 1 143 6",
" 1 143 6", " 1 143 10", " 1 149 4", " 1 149 4",
" 1 15 7", " 1 15 18", " 1 15 21", " 1 15 25",
" 1 15 27", " 1 15 27", " 1 156 2", " 1 156 8",
" 1 21 2", " 1 21 7", " 1 3 4", " 1 3 4",
" 1 33 3", " 1 39 11", " 1 46 3", " 1 52 2",
" 1 52 5", " 1 52 5", " 1 58 4", " 1 64 2",
" 1 64 3", " 1 64 8", " 1 64 8", " 1 76 22",
" 1 76 25", " 1 76 29", " 1 76 29", " 1 82 7",
" 1 82 7", " 1 82 15", " 1 82 21", " 1 82 21"
), v021 = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1), v022 = c(2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2), v023 = c(2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2), bidx = c(1, 2,
1, 2, 1, 2, 3, 1, 1, 1, 1, 2, 3, 1, 2, 1, 1, 2, 1, 1, 1, 1, 1,
2, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 2, 1, 1, 1,
2, 1, 2, 1, 1, 2), b9 = c(0, NA, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, NA, 0, 0, 0, 0, 0), age = c(30,
40, 2, 41, 3, 14, 41, 21, 19, 5, 1, 27, 57, 10, 59, 16, 7, 53,
29, 50, 33, 5, 2, 35, 0, 10, 31, 22, 0, 55, 14, 17, 1, 34, 23,
54, 6, 55, 0, 11, 45, 43, 29, 43, 57, 5, 44, 27, 5, 39)), row.names = c(NA,
50L), class = "data.frame")
#create subset of KRfile to select for children under age of 24 months living with their mother
KRiycf <- KRdata %>%
subset(age < 24 & b9==0 & !is.na(b9)) %>% # children under 24 months living at home
arrange(caseid, bidx) %>% # make sure the data is sorted
subset(is.na(lag(caseid)) | caseid!=lag(caseid)) # select just the youngest
#Above code gives me an empty dataframe
我尝试了几件事
KRiycf <- KRdata %>%
subset(age < 24 & b9 == 0) %>%
arrange(caseid, bidx) %>%
subset(is.na(lag(caseid, default = NA)) | caseid != lag(caseid, default = NA))
##If lag() not returning NA properly, it could be an issue so I tried this but did not work.
KRiycf <- KRdata %>%
subset(age < 24 & b9==0 & !is.na(b9)) %>%
arrange(caseid, bidx) %>%
subset(is.na(lag(caseid)) | caseid!=lag(caseid))
#I thought if there was a NA value in b9 there could be issue with sorting, but that di also not work.
如果您的目标是在
caseidlag()dplyr::slice_min()caseidlibrary(dplyr)
KRdata %>%
subset(age < 24 & b9 %in% 0) %>%
slice_min(age, by = caseid)
输出：
v001 v002 caseid v021 v022 v023 bidx b9 age
1 1 107 1 107 10 1 2 2 1 0 2
2 1 113 1 113 2 1 2 2 1 0 3
3 1 119 1 119 3 1 2 2 1 0 21
4 1 125 1 125 2 1 2 2 1 0 19
5 1 131 1 131 2 1 2 2 1 0 5
6 1 137 1 137 2 1 2 2 1 0 1
7 1 143 1 143 6 1 2 2 1 0 10
8 1 143 1 143 10 1 2 2 1 0 16
9 1 149 1 149 4 1 2 2 1 0 7
10 1 15 1 15 25 1 2 2 1 0 5
11 1 15 1 15 27 1 2 2 1 0 2
12 1 156 1 156 2 1 2 2 1 0 0
13 1 156 1 156 8 1 2 2 1 0 10
14 1 21 1 21 7 1 2 2 1 0 22
15 1 3 1 3 4 1 2 2 1 0 0
16 1 33 1 33 3 1 2 2 1 0 14
17 1 39 1 39 11 1 2 2 1 0 17
18 1 46 1 46 3 1 2 2 1 0 1
19 1 52 1 52 5 1 2 2 1 0 23
20 1 58 1 58 4 1 2 2 1 0 6
21 1 64 1 64 3 1 2 2 1 0 0
22 1 64 1 64 8 1 2 2 1 0 11
23 1 82 1 82 7 1 2 2 1 0 5
24 1 82 1 82 21 1 2 2 1 0 5
您没有提供这些数据所需的输出，所以如果这不完全正确，请告诉我，我会更新！