我可以在 rollapply 中使用什么函数来检查 window 中每个值的条件？
What function can I use in rollapply to check a condition for each value in the window?
我有一个 data.table 与下面的相似。 Returns 这里是完全编造的，但显示数据的一般格式。
> print(dt, n=144)
ID Period Assets Returns
1: a 200601 0 <NA>
2: a 200602 0 <NA>
3: a 200603 0 <NA>
4: a 200604 40000000 <NA>
5: a 200605 45000000 0.3
6: a 200606 48000000 1.43
7: a 200607 52000000 1.54
8: a 200608 55000000 0.09
9: a 200609 57000000 -0.88
10: a 200610 49000000 -0.67
11: a 200611 16000000 2.13
12: a 200612 15000000 2.11
13: a 200701 27000000 -0.9
14: a 200702 90000000 -0.85
15: a 200703 85000000 1.22
16: a 200704 58000000 3.22
17: a 200705 24000000 1.86
18: a 200706 16000000 1.55
19: a 200707 57000000 0.91
20: a 200708 57000000 0.03
21: a 200709 59000000 -0.05
22: a 200710 57450000 -2.55
23: a 200711 56000000 2.13
24: a 200712 65000000 3.16
25: a 200801 57000000 0.91
26: a 200802 65000000 0.56
27: a 200803 70000000 0.64
28: a 200804 70000000 0.43
29: a 200805 78000000 0.99
30: a 200806 43000000 -0.04
31: a 200807 56000000 1
32: a 200808 33000000 1.87
33: a 200809 23000000 -0.33
34: a 200810 21000000 -1.98
35: a 200811 24000000 -1.05
36: a 200812 23000000 0.2
37: a 200901 77000000 0.65
38: a 200902 78000000 0.66
39: a 200903 65000000 0.45
40: a 200904 40000000 1.33
41: a 200905 45000000 -0.93
42: a 200906 48000000 -2.33
43: a 200907 52000000 -0.65
44: a 200908 55000000 -1.23
45: a 200909 57000000 0.02
46: a 200910 49000000 0.22
47: a 200911 56000000 0.45
48: a 200912 45000000 0.19
49: a 201001 27000000 -0.92
50: a 201002 40000000 -1.78
51: a 201003 45000000 0.22
52: a 201004 58000000 0.34
53: a 201005 24000000 2.12
54: a 201006 16000000 0.94
55: a 201007 47000000 0.46
56: a 201008 57000000 -0.04
57: a 201009 59000000 0.67
58: a 201010 57450000 0.19
59: a 201011 56000000 0.34
60: a 201012 65000000 4.12
61: a 201101 57000000 2.98
62: a 201102 65000000 0.69
63: a 201103 70000000 1.21
64: a 201104 70000000 0.96
65: a 201105 0 <NA>
66: a 201106 0 <NA>
67: a 201107 0 <NA>
68: a 201108 0 <NA>
69: a 201109 0 <NA>
70: a 201110 0 <NA>
71: a 201111 0 <NA>
72: a 201112 0 <NA>
73: b 200601 29000000 <NA>
74: b 200602 40000000 1.34
75: b 200603 45000000 2.13
76: b 200604 40000000 0.97
77: b 200605 45000000 0.3
78: b 200606 48000000 1.43
79: b 200607 52000000 1.54
80: b 200608 55000000 0.09
81: b 200609 57000000 -0.88
82: b 200610 49000000 -0.67
83: b 200611 16000000 2.13
84: b 200612 15000000 2.11
85: b 200701 27000000 -0.9
86: b 200702 90000000 -0.85
87: b 200703 85000000 1.22
88: b 200704 58000000 3.22
89: b 200705 24000000 1.86
90: b 200706 16000000 1.55
91: b 200707 57000000 0.91
92: b 200708 57000000 0.03
93: b 200709 59000000 -0.05
94: b 200710 57450000 -2.55
95: b 200711 56000000 2.13
96: b 200712 65000000 3.16
97: b 200801 57000000 0.91
98: b 200802 65000000 0.56
99: b 200803 70000000 0.64
100: b 200804 70000000 0.43
101: b 200805 78000000 0.99
102: b 200806 43000000 -0.04
103: b 200807 56000000 1
104: b 200808 33000000 1.87
105: b 200809 23000000 -0.33
106: b 200810 21000000 -1.98
107: b 200811 24000000 -1.05
108: b 200812 23000000 0.2
109: b 200901 46000000 0.65
110: b 200902 47000000 0.66
111: b 200903 48000000 0.45
112: b 200904 40000000 1.33
113: b 200905 45000000 -0.93
114: b 200906 48000000 -2.33
115: b 200907 52000000 -0.65
116: b 200908 55000000 -1.23
117: b 200909 57000000 0.02
118: b 200910 49000000 0.22
119: b 200911 16000000 0.45
120: b 200912 15000000 0.19
121: b 201001 27000000 -0.92
122: b 201002 90000000 -1.78
123: b 201003 85000000 0.22
124: b 201004 58000000 0.34
125: b 201005 24000000 2.12
126: b 201006 16000000 0.94
127: b 201007 57000000 0.46
128: b 201008 57000000 -0.04
129: b 201009 59000000 0.67
130: b 201010 57450000 0.19
131: b 201011 56000000 0.34
132: b 201012 65000000 4.12
133: b 201101 57000000 2.98
134: b 201102 65000000 0.69
135: b 201103 70000000 1.21
136: b 201104 70000000 0.96
137: b 201105 78000000 -0.12
138: b 201106 43000000 -0.91
139: b 201107 56000000 1.27
140: b 201108 33000000 0.45
141: b 201109 23000000 0.94
142: b 201110 21000000 0.09
143: b 201111 24000000 -0.92
144: b 201112 23000000 -0.92
我想创建一个新列，根据每个 ID 检查前 36 个期间（包括当前期间）中的所有 Returns 值是否不为 NA。我看到的大多数 rollapply 的例子都是为了求和或求标准差。我不想对 window 中的值执行这样的操作；我只是想看看他们是否都满足不NA的条件
这是我的想法，但我不确定我可以使用什么功能以及语法是否正确：
dt[, check1 := all(!is.na(rollapplyr(dt[,Returns], width=36,.N, na.pad=TRUE))), by=ID]
这里是要使用的数据：
library(data.table)
library(zoo)
ID <- c("a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a",
"b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b")
Period <- c(200601L, 200602L, 200603L, 200604L, 200605L, 200606L, 200607L, 200608L, 200609L, 200610L, 200611L, 200612L,
200701L, 200702L, 200703L, 200704L, 200705L, 200706L, 200707L, 200708L, 200709L, 200710L, 200711L, 200712L,
200801L, 200802L, 200803L, 200804L, 200805L, 200806L, 200807L, 200808L, 200809L, 200810L, 200811L, 200812L,
200901L, 200902L, 200903L, 200904L, 200905L, 200906L, 200907L, 200908L, 200909L, 200910L, 200911L, 200912L,
201001L, 201002L, 201003L, 201004L, 201005L, 201006L, 201007L, 201008L, 201009L, 201010L, 201011L, 201012L,
201101L, 201102L, 201103L, 201104L, 201105L, 201106L, 201107L, 201108L, 201109L, 201110L, 201111L, 201112L,
200601L, 200602L, 200603L, 200604L, 200605L, 200606L, 200607L, 200608L, 200609L, 200610L, 200611L, 200612L,
200701L, 200702L, 200703L, 200704L, 200705L, 200706L, 200707L, 200708L, 200709L, 200710L, 200711L, 200712L,
200801L, 200802L, 200803L, 200804L, 200805L, 200806L, 200807L, 200808L, 200809L, 200810L, 200811L, 200812L,
200901L, 200902L, 200903L, 200904L, 200905L, 200906L, 200907L, 200908L, 200909L, 200910L, 200911L, 200912L,
201001L, 201002L, 201003L, 201004L, 201005L, 201006L, 201007L, 201008L, 201009L, 201010L, 201011L, 201012L,
201101L, 201102L, 201103L, 201104L, 201105L, 201106L, 201107L, 201108L, 201109L, 201110L, 201111L, 201112L)
Assets <- c(0L, 0L, 0L, 40000000L, 45000000L, 48000000L, 52000000L, 55000000L, 57000000L, 49000000L,16000000L,15000000L, 27000000L,90000000L,85000000L, 58000000L, 24000000L, 16000000L, 57000000L, 57000000L, 59000000L, 57450000L, 56000000L, 65000000L, 57000000L, 65000000L, 70000000L, 70000000L, 78000000L, 43000000L, 56000000L, 33000000L, 23000000L, 21000000L, 24000000L, 23000000L,
77000000L, 78000000L, 65000000L, 40000000L, 45000000L, 48000000L, 52000000L, 55000000L, 57000000L, 49000000L,56000000L,45000000L, 27000000L,40000000L,45000000L, 58000000L, 24000000L, 16000000L, 47000000L, 57000000L, 59000000L, 57450000L, 56000000L, 65000000L, 57000000L, 65000000L, 70000000L, 70000000L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L,
29000000L, 40000000L, 45000000L, 40000000L, 45000000L, 48000000L, 52000000L, 55000000L, 57000000L, 49000000L,16000000L,15000000L, 27000000L,90000000L,85000000L, 58000000L, 24000000L, 16000000L, 57000000L, 57000000L, 59000000L, 57450000L, 56000000L, 65000000L, 57000000L, 65000000L, 70000000L, 70000000L, 78000000L, 43000000L, 56000000L, 33000000L, 23000000L, 21000000L, 24000000L, 23000000L,
46000000L, 47000000L, 48000000L, 40000000L, 45000000L, 48000000L, 52000000L, 55000000L, 57000000L, 49000000L,16000000L,15000000L, 27000000L,90000000L,85000000L, 58000000L, 24000000L, 16000000L, 57000000L, 57000000L, 59000000L, 57450000L, 56000000L, 65000000L, 57000000L, 65000000L, 70000000L, 70000000L, 78000000L, 43000000L, 56000000L, 33000000L, 23000000L, 21000000L, 24000000L, 23000000L)
Returns <- c(NA, NA, NA, NA, 0.30L, 1.43L, 1.54L, 0.09L, -0.88L, -0.67L, 2.13L, 2.11L, -0.90L, -0.85L, 1.22L, 3.22L, 1.86L, 1.55L, 0.91L, 0.03L, -0.05L, -2.55L, 2.13L, 3.16L, 0.91L, 0.56L, 0.64L, 0.43L, 0.99L, -0.04L, 1.00L, 1.87L, -0.33L, -1.98L, -1.05L, 0.20L,
0.65L, 0.66L, 0.45L, 1.33L, -0.93L, -2.33L, -0.65L, -1.23L, 0.02L, 0.22L, 0.45L, 0.19L, -0.92L, -1.78L, 0.22L, 0.34L, 2.12L, 0.94L, 0.46L, -0.04L, 0.67L, 0.19L, 0.34L, 4.12L, 2.98L, 0.69L, 1.21L, 0.96L, NA, NA, NA, NA, NA, NA, NA, NA,
NA, 1.34L, 2.13L, 0.97L, 0.30L, 1.43L, 1.54L, 0.09L, -0.88L, -0.67L, 2.13L, 2.11L, -0.90L, -0.85L, 1.22L, 3.22L, 1.86L, 1.55L, 0.91L, 0.03L, -0.05L, -2.55L, 2.13L, 3.16L, 0.91L, 0.56L, 0.64L, 0.43L, 0.99L, -0.04L, 1.00L, 1.87L, -0.33L, -1.98L, -1.05L, 0.20L,
0.65L, 0.66L, 0.45L, 1.33L, -0.93L, -2.33L, -0.65L, -1.23L, 0.02L, 0.22L, 0.45L, 0.19L, -0.92L, -1.78L, 0.22L, 0.34L, 2.12L, 0.94L, 0.46L, -0.04L, 0.67L, 0.19L, 0.34L, 4.12L, 2.98L, 0.69L, 1.21L, 0.96L, -0.12L, -0.91L, 1.27L, 0.45L, 0.94L, 0.09L, -0.92L, -0.92L)
dt <- data.table(cbind(ID,Period, Assets, Returns))
谢谢！
你快到了。您需要将 any(is.na(x)) 作为匿名函数传递给 rollapplyr:
dt[, check1 := rollapplyr(Returns, width = 36, function(x) any(is.na(x)), fill=TRUE), by=ID]
dt
#> ID Period Assets Returns check1
#> 1: a 200601 0 <NA> TRUE
#> 2: a 200602 0 <NA> TRUE
#> 3: a 200603 0 <NA> TRUE
#> 4: a 200604 40000000 <NA> TRUE
#> 5: a 200605 45000000 0.3 TRUE
#> ---
#> 140: b 201108 33000000 0.45 FALSE
#> 141: b 201109 23000000 0.94 FALSE
#> 142: b 201110 21000000 0.09 FALSE
#> 143: b 201111 24000000 -0.92 FALSE
#> 144: b 201112 23000000 -0.92 FALSE
或者，更简洁（正如 G Grothendieck 所指出的）
dt[, check1 := rollapplyr(Returns, width = 36, anyNA, fill=TRUE), by=ID]
我有一个 data.table 与下面的相似。 Returns 这里是完全编造的，但显示数据的一般格式。
> print(dt, n=144)
ID Period Assets Returns
1: a 200601 0 <NA>
2: a 200602 0 <NA>
3: a 200603 0 <NA>
4: a 200604 40000000 <NA>
5: a 200605 45000000 0.3
6: a 200606 48000000 1.43
7: a 200607 52000000 1.54
8: a 200608 55000000 0.09
9: a 200609 57000000 -0.88
10: a 200610 49000000 -0.67
11: a 200611 16000000 2.13
12: a 200612 15000000 2.11
13: a 200701 27000000 -0.9
14: a 200702 90000000 -0.85
15: a 200703 85000000 1.22
16: a 200704 58000000 3.22
17: a 200705 24000000 1.86
18: a 200706 16000000 1.55
19: a 200707 57000000 0.91
20: a 200708 57000000 0.03
21: a 200709 59000000 -0.05
22: a 200710 57450000 -2.55
23: a 200711 56000000 2.13
24: a 200712 65000000 3.16
25: a 200801 57000000 0.91
26: a 200802 65000000 0.56
27: a 200803 70000000 0.64
28: a 200804 70000000 0.43
29: a 200805 78000000 0.99
30: a 200806 43000000 -0.04
31: a 200807 56000000 1
32: a 200808 33000000 1.87
33: a 200809 23000000 -0.33
34: a 200810 21000000 -1.98
35: a 200811 24000000 -1.05
36: a 200812 23000000 0.2
37: a 200901 77000000 0.65
38: a 200902 78000000 0.66
39: a 200903 65000000 0.45
40: a 200904 40000000 1.33
41: a 200905 45000000 -0.93
42: a 200906 48000000 -2.33
43: a 200907 52000000 -0.65
44: a 200908 55000000 -1.23
45: a 200909 57000000 0.02
46: a 200910 49000000 0.22
47: a 200911 56000000 0.45
48: a 200912 45000000 0.19
49: a 201001 27000000 -0.92
50: a 201002 40000000 -1.78
51: a 201003 45000000 0.22
52: a 201004 58000000 0.34
53: a 201005 24000000 2.12
54: a 201006 16000000 0.94
55: a 201007 47000000 0.46
56: a 201008 57000000 -0.04
57: a 201009 59000000 0.67
58: a 201010 57450000 0.19
59: a 201011 56000000 0.34
60: a 201012 65000000 4.12
61: a 201101 57000000 2.98
62: a 201102 65000000 0.69
63: a 201103 70000000 1.21
64: a 201104 70000000 0.96
65: a 201105 0 <NA>
66: a 201106 0 <NA>
67: a 201107 0 <NA>
68: a 201108 0 <NA>
69: a 201109 0 <NA>
70: a 201110 0 <NA>
71: a 201111 0 <NA>
72: a 201112 0 <NA>
73: b 200601 29000000 <NA>
74: b 200602 40000000 1.34
75: b 200603 45000000 2.13
76: b 200604 40000000 0.97
77: b 200605 45000000 0.3
78: b 200606 48000000 1.43
79: b 200607 52000000 1.54
80: b 200608 55000000 0.09
81: b 200609 57000000 -0.88
82: b 200610 49000000 -0.67
83: b 200611 16000000 2.13
84: b 200612 15000000 2.11
85: b 200701 27000000 -0.9
86: b 200702 90000000 -0.85
87: b 200703 85000000 1.22
88: b 200704 58000000 3.22
89: b 200705 24000000 1.86
90: b 200706 16000000 1.55
91: b 200707 57000000 0.91
92: b 200708 57000000 0.03
93: b 200709 59000000 -0.05
94: b 200710 57450000 -2.55
95: b 200711 56000000 2.13
96: b 200712 65000000 3.16
97: b 200801 57000000 0.91
98: b 200802 65000000 0.56
99: b 200803 70000000 0.64
100: b 200804 70000000 0.43
101: b 200805 78000000 0.99
102: b 200806 43000000 -0.04
103: b 200807 56000000 1
104: b 200808 33000000 1.87
105: b 200809 23000000 -0.33
106: b 200810 21000000 -1.98
107: b 200811 24000000 -1.05
108: b 200812 23000000 0.2
109: b 200901 46000000 0.65
110: b 200902 47000000 0.66
111: b 200903 48000000 0.45
112: b 200904 40000000 1.33
113: b 200905 45000000 -0.93
114: b 200906 48000000 -2.33
115: b 200907 52000000 -0.65
116: b 200908 55000000 -1.23
117: b 200909 57000000 0.02
118: b 200910 49000000 0.22
119: b 200911 16000000 0.45
120: b 200912 15000000 0.19
121: b 201001 27000000 -0.92
122: b 201002 90000000 -1.78
123: b 201003 85000000 0.22
124: b 201004 58000000 0.34
125: b 201005 24000000 2.12
126: b 201006 16000000 0.94
127: b 201007 57000000 0.46
128: b 201008 57000000 -0.04
129: b 201009 59000000 0.67
130: b 201010 57450000 0.19
131: b 201011 56000000 0.34
132: b 201012 65000000 4.12
133: b 201101 57000000 2.98
134: b 201102 65000000 0.69
135: b 201103 70000000 1.21
136: b 201104 70000000 0.96
137: b 201105 78000000 -0.12
138: b 201106 43000000 -0.91
139: b 201107 56000000 1.27
140: b 201108 33000000 0.45
141: b 201109 23000000 0.94
142: b 201110 21000000 0.09
143: b 201111 24000000 -0.92
144: b 201112 23000000 -0.92
我想创建一个新列，根据每个 ID 检查前 36 个期间（包括当前期间）中的所有 Returns 值是否不为 NA。我看到的大多数 rollapply 的例子都是为了求和或求标准差。我不想对 window 中的值执行这样的操作；我只是想看看他们是否都满足不NA的条件
这是我的想法，但我不确定我可以使用什么功能以及语法是否正确：
dt[, check1 := all(!is.na(rollapplyr(dt[,Returns], width=36,.N, na.pad=TRUE))), by=ID]
这里是要使用的数据：
library(data.table)
library(zoo)
ID <- c("a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a", "a",
"b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b", "b")
Period <- c(200601L, 200602L, 200603L, 200604L, 200605L, 200606L, 200607L, 200608L, 200609L, 200610L, 200611L, 200612L,
200701L, 200702L, 200703L, 200704L, 200705L, 200706L, 200707L, 200708L, 200709L, 200710L, 200711L, 200712L,
200801L, 200802L, 200803L, 200804L, 200805L, 200806L, 200807L, 200808L, 200809L, 200810L, 200811L, 200812L,
200901L, 200902L, 200903L, 200904L, 200905L, 200906L, 200907L, 200908L, 200909L, 200910L, 200911L, 200912L,
201001L, 201002L, 201003L, 201004L, 201005L, 201006L, 201007L, 201008L, 201009L, 201010L, 201011L, 201012L,
201101L, 201102L, 201103L, 201104L, 201105L, 201106L, 201107L, 201108L, 201109L, 201110L, 201111L, 201112L,
200601L, 200602L, 200603L, 200604L, 200605L, 200606L, 200607L, 200608L, 200609L, 200610L, 200611L, 200612L,
200701L, 200702L, 200703L, 200704L, 200705L, 200706L, 200707L, 200708L, 200709L, 200710L, 200711L, 200712L,
200801L, 200802L, 200803L, 200804L, 200805L, 200806L, 200807L, 200808L, 200809L, 200810L, 200811L, 200812L,
200901L, 200902L, 200903L, 200904L, 200905L, 200906L, 200907L, 200908L, 200909L, 200910L, 200911L, 200912L,
201001L, 201002L, 201003L, 201004L, 201005L, 201006L, 201007L, 201008L, 201009L, 201010L, 201011L, 201012L,
201101L, 201102L, 201103L, 201104L, 201105L, 201106L, 201107L, 201108L, 201109L, 201110L, 201111L, 201112L)
Assets <- c(0L, 0L, 0L, 40000000L, 45000000L, 48000000L, 52000000L, 55000000L, 57000000L, 49000000L,16000000L,15000000L, 27000000L,90000000L,85000000L, 58000000L, 24000000L, 16000000L, 57000000L, 57000000L, 59000000L, 57450000L, 56000000L, 65000000L, 57000000L, 65000000L, 70000000L, 70000000L, 78000000L, 43000000L, 56000000L, 33000000L, 23000000L, 21000000L, 24000000L, 23000000L,
77000000L, 78000000L, 65000000L, 40000000L, 45000000L, 48000000L, 52000000L, 55000000L, 57000000L, 49000000L,56000000L,45000000L, 27000000L,40000000L,45000000L, 58000000L, 24000000L, 16000000L, 47000000L, 57000000L, 59000000L, 57450000L, 56000000L, 65000000L, 57000000L, 65000000L, 70000000L, 70000000L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L,
29000000L, 40000000L, 45000000L, 40000000L, 45000000L, 48000000L, 52000000L, 55000000L, 57000000L, 49000000L,16000000L,15000000L, 27000000L,90000000L,85000000L, 58000000L, 24000000L, 16000000L, 57000000L, 57000000L, 59000000L, 57450000L, 56000000L, 65000000L, 57000000L, 65000000L, 70000000L, 70000000L, 78000000L, 43000000L, 56000000L, 33000000L, 23000000L, 21000000L, 24000000L, 23000000L,
46000000L, 47000000L, 48000000L, 40000000L, 45000000L, 48000000L, 52000000L, 55000000L, 57000000L, 49000000L,16000000L,15000000L, 27000000L,90000000L,85000000L, 58000000L, 24000000L, 16000000L, 57000000L, 57000000L, 59000000L, 57450000L, 56000000L, 65000000L, 57000000L, 65000000L, 70000000L, 70000000L, 78000000L, 43000000L, 56000000L, 33000000L, 23000000L, 21000000L, 24000000L, 23000000L)
Returns <- c(NA, NA, NA, NA, 0.30L, 1.43L, 1.54L, 0.09L, -0.88L, -0.67L, 2.13L, 2.11L, -0.90L, -0.85L, 1.22L, 3.22L, 1.86L, 1.55L, 0.91L, 0.03L, -0.05L, -2.55L, 2.13L, 3.16L, 0.91L, 0.56L, 0.64L, 0.43L, 0.99L, -0.04L, 1.00L, 1.87L, -0.33L, -1.98L, -1.05L, 0.20L,
0.65L, 0.66L, 0.45L, 1.33L, -0.93L, -2.33L, -0.65L, -1.23L, 0.02L, 0.22L, 0.45L, 0.19L, -0.92L, -1.78L, 0.22L, 0.34L, 2.12L, 0.94L, 0.46L, -0.04L, 0.67L, 0.19L, 0.34L, 4.12L, 2.98L, 0.69L, 1.21L, 0.96L, NA, NA, NA, NA, NA, NA, NA, NA,
NA, 1.34L, 2.13L, 0.97L, 0.30L, 1.43L, 1.54L, 0.09L, -0.88L, -0.67L, 2.13L, 2.11L, -0.90L, -0.85L, 1.22L, 3.22L, 1.86L, 1.55L, 0.91L, 0.03L, -0.05L, -2.55L, 2.13L, 3.16L, 0.91L, 0.56L, 0.64L, 0.43L, 0.99L, -0.04L, 1.00L, 1.87L, -0.33L, -1.98L, -1.05L, 0.20L,
0.65L, 0.66L, 0.45L, 1.33L, -0.93L, -2.33L, -0.65L, -1.23L, 0.02L, 0.22L, 0.45L, 0.19L, -0.92L, -1.78L, 0.22L, 0.34L, 2.12L, 0.94L, 0.46L, -0.04L, 0.67L, 0.19L, 0.34L, 4.12L, 2.98L, 0.69L, 1.21L, 0.96L, -0.12L, -0.91L, 1.27L, 0.45L, 0.94L, 0.09L, -0.92L, -0.92L)
dt <- data.table(cbind(ID,Period, Assets, Returns))
谢谢！
你快到了。您需要将 any(is.na(x)) 作为匿名函数传递给 rollapplyr:
dt[, check1 := rollapplyr(Returns, width = 36, function(x) any(is.na(x)), fill=TRUE), by=ID]
dt
#> ID Period Assets Returns check1
#> 1: a 200601 0 <NA> TRUE
#> 2: a 200602 0 <NA> TRUE
#> 3: a 200603 0 <NA> TRUE
#> 4: a 200604 40000000 <NA> TRUE
#> 5: a 200605 45000000 0.3 TRUE
#> ---
#> 140: b 201108 33000000 0.45 FALSE
#> 141: b 201109 23000000 0.94 FALSE
#> 142: b 201110 21000000 0.09 FALSE
#> 143: b 201111 24000000 -0.92 FALSE
#> 144: b 201112 23000000 -0.92 FALSE
或者，更简洁（正如 G Grothendieck 所指出的）
dt[, check1 := rollapplyr(Returns, width = 36, anyNA, fill=TRUE), by=ID]