通过将所有预先存在的变量除以所有其他变量来创建新变量
Create new variables by dividing all pre-exisiting variables by all other variables
我想通过将所有预先存在的变量彼此相除来创建新变量
例如
X1/X1、X1/X2、X1/X3、X1/X4、X1/X5、X1/X6、X1/X7、X1/X8、X1/X9, X1/X10,
X2/X1、X2/X2、X2/X3、X2/X4、X2/X5、X2/X6、X2/X7、X2/X8、X2/X9, X2/X10,
X3/X1、X3/X2 ...
我开始尝试单独执行每个操作，如下所示，但我需要用多个变量名复制它，因此自动化（我假设 function/lapply）将是理想的。
ds$rom_3_5m <- (ds$roll_open_mean_3m/ds$roll_open_mean_5m)
ds$rom_3_10m <- (ds$roll_open_mean_3m/ds$roll_open_mean_10m)
ds$rom_3_15m <- (ds$roll_open_mean_3m/ds$roll_open_mean_15m)
ds$rom_3_30m <- (ds$roll_open_mean_3m/ds$roll_open_mean_30m)
ds$rom_3_60m <- (ds$roll_open_mean_3m/ds$roll_open_mean_60m)
ds$rom_3_120m <- (ds$roll_open_mean_3m/ds$roll_open_mean_120m)
ds$rom_3_240m <- (ds$roll_open_mean_3m/ds$roll_open_mean_240m)
ds$rom_3_480m <- (ds$roll_open_mean_3m/ds$roll_open_mean_480m)
ds$rom_3_960m <- (ds$roll_open_mean_3m/ds$roll_open_mean_960m)
ds$rom_3_1920m <- (ds$roll_open_mean_3m/ds$roll_open_mean_1920m)
ds$rom_3_3840m <- (ds$roll_open_mean_3m/ds$roll_open_mean_3840m)
ds$rom_3_7680m <- (ds$roll_open_mean_3m/ds$roll_open_mean_7680m)
ds$rom_3_15360m <- (ds$roll_open_mean_3m/ds$roll_open_mean_15360m)
ds$rom_3_30720m <- (ds$roll_open_mean_3m/ds$roll_open_mean_30720m)
ds$rom_3_61440m <- (ds$roll_open_mean_3m/ds$roll_open_mean_61440m)
ds$rom_3_122880m <- (ds$roll_open_mean_3m/ds$roll_open_mean_122880m)
ds$rom_3_245760m <- (ds$roll_open_mean_3m/ds$roll_open_mean_245760m)
ds$rom_3_491520m <- (ds$roll_open_mean_3m/ds$roll_open_mean_491520m)
#5m
ds$rom_5_3m <- (ds$roll_open_mean_5m/ds$roll_open_mean_3m)
ds$rom_5_10m <- (ds$roll_open_mean_5m/ds$roll_open_mean_10m)
ds$rom_5_15m <- (ds$roll_open_mean_5m/ds$roll_open_mean_15m)
ds$rom_5_30m <- (ds$roll_open_mean_5m/ds$roll_open_mean_30m)
ds$rom_5_60m <- (ds$roll_open_mean_5m/ds$roll_open_mean_60m)
ds$rom_5_120m <- (ds$roll_open_mean_5m/ds$roll_open_mean_120m)
ds$rom_5_240m <- (ds$roll_open_mean_5m/ds$roll_open_mean_240m)
ds$rom_5_480m <- (ds$roll_open_mean_5m/ds$roll_open_mean_480m)
ds$rom_5_960m <- (ds$roll_open_mean_5m/ds$roll_open_mean_960m)
ds$rom_5_1920m <- (ds$roll_open_mean_5m/ds$roll_open_mean_1920m)
ds$rom_5_3840m <- (ds$roll_open_mean_5m/ds$roll_open_mean_3840m)
ds$rom_5_7680m <- (ds$roll_open_mean_5m/ds$roll_open_mean_7680m)
ds$rom_5_15360m <- (ds$roll_open_mean_5m/ds$roll_open_mean_15360m)
ds$rom_5_30720m <- (ds$roll_open_mean_5m/ds$roll_open_mean_30720m)
ds$rom_5_61440m <- (ds$roll_open_mean_5m/ds$roll_open_mean_61440m)
ds$rom_5_122880m <- (ds$roll_open_mean_5m/ds$roll_open_mean_122880m)
ds$rom_5_245760m <- (ds$roll_open_mean_5m/ds$roll_open_mean_245760m)
ds$rom_5_491520m <- (ds$roll_open_mean_5m/ds$roll_open_mean_491520m)
#10m
ds$rom_10_3m <- (ds$roll_open_mean_10m/ds$roll_open_mean_3m)
ds$rom_10_5m <- (ds$roll_open_mean_10m/ds$roll_open_mean_5m)
ds$rom_10_15m <- (ds$roll_open_mean_10m/ds$roll_open_mean_15m)
我有一个包含 40 多个变量和 600 万行的数据框，我在下面附上了一个较小的示例数据框。
提前致谢！
查理
X1 X2 X3 X4 X5 X6 X7 X8 X9 X10
1 57 77 48 8 31 43 47 13 26 88
2 25 75 86 77 4 65 5 49 31 57
3 91 90 42 69 82 33 56 99 47 39
4 35 96 86 77 67 77 20 17 77 92
5 6 100 50 62 16 31 0 39 72 4
6 90 34 74 89 71 37 73 45 24 28
7 24 22 92 13 57 97 32 2 12 80
8 74 59 49 2 97 100 15 37 15 67
9 43 38 66 97 8 20 85 25 97 67
10 82 4 56 40 42 46 44 98 98 76
11 60 68 92 99 81 92 78 59 23 81
12 22 57 37 100 7 1 89 41 40 56
13 69 13 1 82 89 45 83 24 71 29
14 8 14 66 48 94 8 20 3 28 63
15 26 70 56 62 9 34 11 86 71 64
16 7 55 15 100 91 89 46 74 98 14
17 29 68 19 66 83 29 84 76 90 45
18 27 76 6 48 17 28 8 7 52 37
19 68 58 51 75 60 57 74 46 98 93
20 15 15 89 55 23 3 3 8 32 37
21 78 49 57 48 96 89 4 95 67 58
22 12 36 42 59 27 92 48 0 92 28
23 51 17 77 61 84 53 46 22 27 36
24 40 84 83 35 19 13 80 78 96 87
25 44 80 25 72 43 17 74 70 52 36
26 14 61 63 82 16 47 32 93 19 84
27 93 19 28 62 74 1 85 65 50 9
28 80 62 6 58 48 97 97 18 65 43
29 12 58 95 79 37 89 89 83 22 85
30 57 73 22 88 99 63 58 87 90 66
正如@27 ϕ 9 在评论中建议的那样，您应该使用 lapply 解决方案。
这样，您还可以创建一个具有正确名称的唯一数据框
l <- lapply(df, `/`, df)
l <- unlist(l, recursive = FALSE)
data.frame(l)
我想通过将所有预先存在的变量彼此相除来创建新变量
例如 X1/X1、X1/X2、X1/X3、X1/X4、X1/X5、X1/X6、X1/X7、X1/X8、X1/X9, X1/X10, X2/X1、X2/X2、X2/X3、X2/X4、X2/X5、X2/X6、X2/X7、X2/X8、X2/X9, X2/X10, X3/X1、X3/X2 ...
我开始尝试单独执行每个操作，如下所示，但我需要用多个变量名复制它，因此自动化（我假设 function/lapply）将是理想的。
ds$rom_3_5m <- (ds$roll_open_mean_3m/ds$roll_open_mean_5m)
ds$rom_3_10m <- (ds$roll_open_mean_3m/ds$roll_open_mean_10m)
ds$rom_3_15m <- (ds$roll_open_mean_3m/ds$roll_open_mean_15m)
ds$rom_3_30m <- (ds$roll_open_mean_3m/ds$roll_open_mean_30m)
ds$rom_3_60m <- (ds$roll_open_mean_3m/ds$roll_open_mean_60m)
ds$rom_3_120m <- (ds$roll_open_mean_3m/ds$roll_open_mean_120m)
ds$rom_3_240m <- (ds$roll_open_mean_3m/ds$roll_open_mean_240m)
ds$rom_3_480m <- (ds$roll_open_mean_3m/ds$roll_open_mean_480m)
ds$rom_3_960m <- (ds$roll_open_mean_3m/ds$roll_open_mean_960m)
ds$rom_3_1920m <- (ds$roll_open_mean_3m/ds$roll_open_mean_1920m)
ds$rom_3_3840m <- (ds$roll_open_mean_3m/ds$roll_open_mean_3840m)
ds$rom_3_7680m <- (ds$roll_open_mean_3m/ds$roll_open_mean_7680m)
ds$rom_3_15360m <- (ds$roll_open_mean_3m/ds$roll_open_mean_15360m)
ds$rom_3_30720m <- (ds$roll_open_mean_3m/ds$roll_open_mean_30720m)
ds$rom_3_61440m <- (ds$roll_open_mean_3m/ds$roll_open_mean_61440m)
ds$rom_3_122880m <- (ds$roll_open_mean_3m/ds$roll_open_mean_122880m)
ds$rom_3_245760m <- (ds$roll_open_mean_3m/ds$roll_open_mean_245760m)
ds$rom_3_491520m <- (ds$roll_open_mean_3m/ds$roll_open_mean_491520m)
#5m
ds$rom_5_3m <- (ds$roll_open_mean_5m/ds$roll_open_mean_3m)
ds$rom_5_10m <- (ds$roll_open_mean_5m/ds$roll_open_mean_10m)
ds$rom_5_15m <- (ds$roll_open_mean_5m/ds$roll_open_mean_15m)
ds$rom_5_30m <- (ds$roll_open_mean_5m/ds$roll_open_mean_30m)
ds$rom_5_60m <- (ds$roll_open_mean_5m/ds$roll_open_mean_60m)
ds$rom_5_120m <- (ds$roll_open_mean_5m/ds$roll_open_mean_120m)
ds$rom_5_240m <- (ds$roll_open_mean_5m/ds$roll_open_mean_240m)
ds$rom_5_480m <- (ds$roll_open_mean_5m/ds$roll_open_mean_480m)
ds$rom_5_960m <- (ds$roll_open_mean_5m/ds$roll_open_mean_960m)
ds$rom_5_1920m <- (ds$roll_open_mean_5m/ds$roll_open_mean_1920m)
ds$rom_5_3840m <- (ds$roll_open_mean_5m/ds$roll_open_mean_3840m)
ds$rom_5_7680m <- (ds$roll_open_mean_5m/ds$roll_open_mean_7680m)
ds$rom_5_15360m <- (ds$roll_open_mean_5m/ds$roll_open_mean_15360m)
ds$rom_5_30720m <- (ds$roll_open_mean_5m/ds$roll_open_mean_30720m)
ds$rom_5_61440m <- (ds$roll_open_mean_5m/ds$roll_open_mean_61440m)
ds$rom_5_122880m <- (ds$roll_open_mean_5m/ds$roll_open_mean_122880m)
ds$rom_5_245760m <- (ds$roll_open_mean_5m/ds$roll_open_mean_245760m)
ds$rom_5_491520m <- (ds$roll_open_mean_5m/ds$roll_open_mean_491520m)
#10m
ds$rom_10_3m <- (ds$roll_open_mean_10m/ds$roll_open_mean_3m)
ds$rom_10_5m <- (ds$roll_open_mean_10m/ds$roll_open_mean_5m)
ds$rom_10_15m <- (ds$roll_open_mean_10m/ds$roll_open_mean_15m)
我有一个包含 40 多个变量和 600 万行的数据框，我在下面附上了一个较小的示例数据框。
提前致谢！
查理
X1 X2 X3 X4 X5 X6 X7 X8 X9 X10
1 57 77 48 8 31 43 47 13 26 88
2 25 75 86 77 4 65 5 49 31 57
3 91 90 42 69 82 33 56 99 47 39
4 35 96 86 77 67 77 20 17 77 92
5 6 100 50 62 16 31 0 39 72 4
6 90 34 74 89 71 37 73 45 24 28
7 24 22 92 13 57 97 32 2 12 80
8 74 59 49 2 97 100 15 37 15 67
9 43 38 66 97 8 20 85 25 97 67
10 82 4 56 40 42 46 44 98 98 76
11 60 68 92 99 81 92 78 59 23 81
12 22 57 37 100 7 1 89 41 40 56
13 69 13 1 82 89 45 83 24 71 29
14 8 14 66 48 94 8 20 3 28 63
15 26 70 56 62 9 34 11 86 71 64
16 7 55 15 100 91 89 46 74 98 14
17 29 68 19 66 83 29 84 76 90 45
18 27 76 6 48 17 28 8 7 52 37
19 68 58 51 75 60 57 74 46 98 93
20 15 15 89 55 23 3 3 8 32 37
21 78 49 57 48 96 89 4 95 67 58
22 12 36 42 59 27 92 48 0 92 28
23 51 17 77 61 84 53 46 22 27 36
24 40 84 83 35 19 13 80 78 96 87
25 44 80 25 72 43 17 74 70 52 36
26 14 61 63 82 16 47 32 93 19 84
27 93 19 28 62 74 1 85 65 50 9
28 80 62 6 58 48 97 97 18 65 43
29 12 58 95 79 37 89 89 83 22 85
30 57 73 22 88 99 63 58 87 90 66
正如@27 ϕ 9 在评论中建议的那样，您应该使用 lapply 解决方案。
这样，您还可以创建一个具有正确名称的唯一数据框
l <- lapply(df, `/`, df)
l <- unlist(l, recursive = FALSE)
data.frame(l)