I must be blind but I don' see any link between the two files. RD2 seems to
give some kind of encounter data in length of time. allocate gives the time
that an animal was wearing a collar ?
If each file belongs to a specific animal (am I reading this correctly) what
does it matter what the collar was? Or is there something in the RD2 file that
gives a clue to other collars ?
I must be missing the point completely but I don't understand why you want
to know who was wearing what at what time from the data you have presented.
BTW , nice presentation of the sample data however dput (head(RD2, 200)) would
have been even better than head(RD2, 200).
John Kane
Kingston ON Canada
> -----Original Message-----
> From: cat.e.cowie at gmail.com
> Sent: Fri, 29 Mar 2013 18:49:45 +0800
> To: r-help at r-project.org
> Subject: [R] Create values based on a table of conditions
>
> Hi R help forum,
>
> I have a simple data frame of four columns - one of numbers (really a
> categorical variable), one of dates and one
>
> of data. I have over 500,000 data points to work with, spread over 40
> files, each named after a different animal.
>
> These are contact data recorded by proximity loggers over two years
> between the animals of the file name and
>
> collars being worn by other animals. The collar may have been worn by
> more than one different animal in the study period.
>
> So for example...
>
> filename: RD2
>
> And the data looks like:
>
> (I'm sorry here, dput(head(RD2, 200)) was outputting all the times,
> such that it was too large for the console, so I just used head() )
>
>> head(RD2, 200)
> encounter_id date encounter_start encounter_length
> 1 15 22/06/2011 22:21:45 20
> 2 10 04/07/2011 04:18:57 1
> 3 10 04/07/2011 04:23:12 1
> 4 15 08/07/2011 02:02:30 23
> 5 15 08/07/2011 02:03:26 264
> 6 15 08/07/2011 02:09:20 35
> 7 15 08/07/2011 02:10:25 173
> 8 15 08/07/2011 02:13:40 65
> 9 15 08/07/2011 02:15:12 47
> 10 15 08/07/2011 02:17:16 65
> 11 15 08/07/2011 02:39:10 54
> 12 15 14/07/2011 02:20:27 1
> 13 15 14/07/2011 02:23:35 6
> 14 15 14/07/2011 02:25:52 1
> 15 15 14/07/2011 02:26:17 39
> 16 15 14/07/2011 02:28:36 31
> 17 60 14/07/2011 02:41:56 1
> 18 60 14/07/2011 02:42:30 1
> 19 60 14/07/2011 02:43:30 36
> 20 60 14/07/2011 02:45:38 1
> 21 60 14/07/2011 02:46:07 15
> 22 60 14/07/2011 02:46:59 11
> 23 52 14/07/2011 02:57:43 97
> 24 12 09/08/2011 22:13:42 1
> 25 41 18/08/2011 02:53:18 1
> 26 41 19/08/2011 00:57:46 1
> 27 41 19/08/2011 01:59:17 1
> 28 41 19/08/2011 02:06:16 21
> 29 41 19/08/2011 02:08:50 1
> 30 41 19/08/2011 02:09:26 29
> 31 50 19/08/2011 02:19:31 1
> 32 50 19/08/2011 02:20:20 1
> 33 50 19/08/2011 02:20:54 1
> 34 50 19/08/2011 02:21:25 25
> 35 50 19/08/2011 02:22:17 20
> 36 50 19/08/2011 02:23:00 10
> 37 50 19/08/2011 02:25:10 1
> 38 50 19/08/2011 02:25:54 1
> 39 50 19/08/2011 02:26:50 75
> 40 50 19/08/2011 02:29:53 1
> 41 50 19/08/2011 02:30:41 6
> 42 50 28/08/2011 19:32:47 1
> 43 50 28/08/2011 19:36:38 10
> 44 39 30/08/2011 22:13:14 1
> 45 52 13/09/2011 20:04:57 17
> 46 20 18/09/2011 00:32:43 8
> 47 17 02/10/2011 09:10:18 52
> 48 17 02/10/2011 09:15:49 1
> 49 17 02/10/2011 09:17:19 1
> 50 25 03/10/2011 17:16:20 6
> 51 41 05/10/2011 04:04:35 1
> 52 52 07/10/2011 22:59:57 45
> 53 62 16/10/2011 18:33:03 6
> 54 52 28/10/2011 15:19:47 1
> 55 41 28/10/2011 19:00:24 8
> 56 17 03/12/2011 17:38:08 2
> 57 17 03/12/2011 17:38:49 20
> 58 17 20/05/2012 19:45:17 17
> 59 17 20/05/2012 19:45:58 47
> 60 17 22/05/2012 23:12:54 1
> 61 17 22/05/2012 23:15:15 3
> 62 17 22/05/2012 23:15:40 7
> 63 17 22/05/2012 23:16:09 40
> 64 17 22/05/2012 23:36:30 3
> 65 17 22/05/2012 23:37:29 10
> 66 17 22/05/2012 23:39:17 1
> 67 17 23/05/2012 20:44:01 15
> 68 6 17/07/2012 04:25:29 8
> 69 53 17/07/2012 23:53:01 2
> 70 53 24/07/2012 03:09:25 1
> 71 17 24/07/2012 20:58:42 16
> 72 17 24/07/2012 21:00:46 13
> 73 17 25/07/2012 04:57:54 21
> 74 17 25/07/2012 05:04:37 1
> 75 17 27/07/2012 00:53:59 125
> 76 17 27/07/2012 01:00:55 1
> 77 17 27/07/2012 01:01:57 1
> 78 17 27/07/2012 01:28:46 1
> 79 17 27/07/2012 01:41:14 5
> 80 17 27/07/2012 01:42:18 1
> 81 17 27/07/2012 01:43:46 1
> 82 17 27/07/2012 01:49:23 4
> 83 17 27/07/2012 04:26:49 1
> 84 17 27/07/2012 04:44:38 3
> 85 17 27/07/2012 04:45:27 1
> 86 17 27/07/2012 05:21:40 2
> 87 17 27/07/2012 05:22:13 2
> 88 17 27/07/2012 05:39:47 1
> 89 17 28/07/2012 00:11:46 3
> 90 17 30/07/2012 20:58:47 1
> 91 17 30/07/2012 20:59:22 1
> 92 17 30/07/2012 20:59:50 1
> 93 17 30/07/2012 21:03:15 1
> 94 17 30/07/2012 21:04:00 10
> 95 17 30/07/2012 21:05:05 1
> 96 17 30/07/2012 21:15:02 1
> 97 17 30/07/2012 22:04:55 1
> 98 17 30/07/2012 22:27:21 1
> 99 17 30/07/2012 22:33:12 5
> 100 17 30/07/2012 22:35:55 3
> 101 17 30/07/2012 22:36:32 4
> 102 17 30/07/2012 22:38:38 37
> 103 17 30/07/2012 22:40:45 6
> 104 17 30/07/2012 22:43:13 2
> 105 17 30/07/2012 22:46:56 3
> 106 17 30/07/2012 22:54:33 29
> 107 17 30/07/2012 23:12:10 1
> 108 17 30/07/2012 23:12:36 25
> 109 17 30/07/2012 23:14:42 1
> 110 17 30/07/2012 23:20:08 1
> 111 17 30/07/2012 23:29:27 11
> 112 17 30/07/2012 23:35:28 1
> 113 17 30/07/2012 23:36:44 1
> 114 17 30/07/2012 23:43:15 1
> 115 17 30/07/2012 23:43:49 1
> 116 17 30/07/2012 23:44:39 29
> 117 17 30/07/2012 23:45:31 1
> 118 17 30/07/2012 23:52:18 23
> 119 17 30/07/2012 23:53:20 2
> 120 17 30/07/2012 23:53:57 12
> 121 17 30/07/2012 23:54:37 19
> 122 17 30/07/2012 23:59:34 19
> 123 17 31/07/2012 00:13:14 5
> 124 17 31/07/2012 00:13:44 8
> 125 17 31/07/2012 00:16:08 11
> 126 17 31/07/2012 00:16:58 39
> 127 17 31/07/2012 00:18:16 13
> 128 17 31/07/2012 00:19:43 12
> 129 17 31/07/2012 00:20:26 7
> 130 17 31/07/2012 00:21:05 1
> 131 17 31/07/2012 00:27:05 4
> 132 17 31/07/2012 00:54:22 1
> 133 17 31/07/2012 00:54:48 96
> 134 17 31/07/2012 01:01:57 1
> 135 17 31/07/2012 01:47:05 1
> 136 17 31/07/2012 01:47:50 20
> 137 17 31/07/2012 01:48:32 8
> 138 17 31/07/2012 03:37:37 1
> 139 17 31/07/2012 03:38:03 5
> 140 17 31/07/2012 03:38:44 99
> 141 17 31/07/2012 03:41:34 28
> 142 17 31/07/2012 03:43:49 22
> 143 17 31/07/2012 03:45:43 1
> 144 17 31/07/2012 03:49:19 24
> 145 17 31/07/2012 04:03:03 6
> 146 17 31/07/2012 04:06:10 65
> 147 17 31/07/2012 04:11:19 1
> 148 17 31/07/2012 04:11:49 1
> 149 17 31/07/2012 04:24:48 3
> 150 17 31/07/2012 04:25:39 80
> 151 17 31/07/2012 04:27:28 17
> 152 17 31/07/2012 04:32:41 7
> 153 17 31/07/2012 04:45:24 1
> 154 17 31/07/2012 04:56:40 6
> 155 17 31/07/2012 05:05:56 2
> 156 17 31/07/2012 05:07:29 1
> 157 17 01/08/2012 00:21:25 1
> 158 17 01/08/2012 00:22:17 25
> 159 17 01/08/2012 00:23:36 12
> 160 17 01/08/2012 00:25:14 4
> 161 17 01/08/2012 00:41:13 1
> 162 17 01/08/2012 00:42:28 12
> 163 17 01/08/2012 00:43:54 2
> 164 17 01/08/2012 00:55:37 8
> 165 17 01/08/2012 01:00:38 39
> 166 17 01/08/2012 01:06:32 9
> 167 17 01/08/2012 01:11:19 1
> 168 17 01/08/2012 01:14:10 3
> 169 17 01/08/2012 01:14:48 17
> 170 17 01/08/2012 01:15:30 27
> 171 17 01/08/2012 01:16:20 71
> 172 17 01/08/2012 01:18:05 107
> 173 17 01/08/2012 01:23:54 1
> 174 17 01/08/2012 01:24:51 11
> 175 17 01/08/2012 01:32:09 1
> 176 17 01/08/2012 01:32:57 16
> 177 17 01/08/2012 01:33:35 69
> 178 17 01/08/2012 01:38:47 1
> 179 17 01/08/2012 01:39:18 75
> 180 17 01/08/2012 01:41:07 1
> 181 17 01/08/2012 01:41:48 1
> 182 17 01/08/2012 01:53:50 1
> 183 17 01/08/2012 01:55:29 1
> 184 17 01/08/2012 01:55:52 2
> 185 17 01/08/2012 01:56:17 25
> 186 17 01/08/2012 02:01:00 3
> 187 17 01/08/2012 02:03:13 32
> 188 17 01/08/2012 02:04:34 99
> 189 17 01/08/2012 02:06:43 43
> 190 17 01/08/2012 02:07:58 7
> 191 17 01/08/2012 02:08:55 22
> 192 17 01/08/2012 02:09:47 6
> 193 17 01/08/2012 02:20:21 15
> 194 17 01/08/2012 04:06:42 1
> 195 17 01/08/2012 04:15:26 46
> 196 17 01/08/2012 04:21:12 1
> 197 17 01/08/2012 04:22:12 1
> 198 17 01/08/2012 04:34:28 25
> 199 17 01/08/2012 04:35:26 1
> 200 17 01/08/2012 04:36:01 14
>
> When an animal was caught and the collar battery needed to be changed,
> a whole new collar had to be put on, as
>
> these animals (cows, pigs, wild boar and red deer!) were not that easy
> to catch, or hold onto for long! In order to follow the movements of
> each animal rather than each collar I now need to create a new column
> that assigns the data by animal rather than by collar. I have a table
> of dates showing which animal was wearing which collar and when:
>
>> head(allocate)
> animal collar start_date end_date
> 1 CO1 36 16/07/2010 11/04/2011
> 2 CO1 64 12/04/2011 21/02/2012
> 3 CO1 25 21/02/2012 10/07/2012
> 4 CO2 28 16/07/2010 09/04/2011
> 5 CO2 11 12/04/2011 10/04/2012
> 6 CO3 46 16/07/2010 10/04/2011
>
>> dput(allocate)
> structure(list(animal = structure(c(1L, 1L, 1L, 10L, 10L, 11L,
> 11L, 11L, 12L, 13L, 13L, 14L, 15L, 16L, 16L, 16L, 17L, 17L, 17L,
> 2L, 3L, 4L, 4L, 4L, 5L, 6L, 6L, 7L, 8L, 9L, 9L, 26L, 26L, 26L,
> 31L, 31L, 31L, 31L, 32L, 32L, 32L, 32L, 33L, 33L, 33L, 34L, 34L,
> 34L, 35L, 35L, 36L, 36L, 37L, 38L, 38L, 27L, 27L, 28L, 28L, 29L,
> 29L, 29L, 30L, 30L, 30L, 18L, 19L, 20L, 21L, 22L, 23L, 24L, 25L,
> 39L, 40L, 41L), .Label = c("CO1", "CO10",
"CO11", "CO12", "CO13",
> "CO14", "CO15", "CO16", "CO17",
"CO2", "CO3", "CO4", "CO5",
"CO6",
> "CO7", "CO8", "CO9", "PI1",
"PI2", "PI3", "PI4", "PI5",
"PI6",
> "PI7", "PI8", "PO1", "PO10",
"PO11", "PO12", "PO13", "PO2",
"PO3",
> "PO4", "PO5", "PO6", "PO7",
"PO8", "PO9", "RD1", "RD2",
"WB1"
> ), class = "factor"), collar = c(36L, 64L, 25L, 28L, 11L, 46L,
> 58L, 12L, 4L, 52L, 2L, 63L, 14L, 12L, 12L, 57L, 51L, 51L, 24L,
> 42L, 44L, 25L, 23L, 7L, 2L, 45L, 9L, 11L, 30L, 14L, 42L, 23L,
> 19L, 52L, 41L, 53L, 18L, 41L, 55L, 10L, 15L, 50L, 34L, 3L, 16L,
> 32L, 16L, 37L, 50L, 19L, 37L, 3L, 10L, 27L, 41L, 6L, 15L, 62L,
> 18L, 52L, 39L, 52L, 62L, 53L, 62L, 18L, 63L, 37L, 57L, 54L, 23L,
> 56L, 38L, 17L, 61L, 33L), start_date = structure(c(14L, 7L, 22L,
> 14L, 7L, 14L, 7L, 3L, 7L, 14L, 13L, 11L, 14L, 14L, 16L, 15L,
> 14L, 13L, 15L, 29L, 14L, 14L, 13L, 15L, 14L, 14L, 15L, 6L, 6L,
> 13L, 12L, 31L, 6L, 18L, 29L, 6L, 25L, 10L, 30L, 6L, 23L, 10L,
> 28L, 4L, 22L, 30L, 6L, 22L, 6L, 20L, 24L, 22L, 27L, 2L, 23L,
> 24L, 10L, 27L, 9L, 6L, 27L, 8L, 6L, 23L, 9L, 17L, 5L, 17L, 1L,
> 17L, 1L, 1L, 17L, 21L, 19L, 26L), .Label = c("02/06/2011",
"02/11/2010",
> "09/02/2012", "10/04/2011", "10/12/2010",
"11/04/2011", "12/04/2011",
> "13/02/2012", "14/02/2012", "15/02/2012",
"15/07/2010", "16/02/2012",
> "16/03/2011", "16/07/2010", "17/02/2012",
"17/03/2011", "18/10/2010",
> "19/07/2011", "20/06/2011", "20/07/2011",
"20/08/2011", "21/02/2012",
> "21/07/2011", "21/09/2011", "22/09/2011",
"23/07/2011", "25/07/2011",
> "25/08/2010", "26/10/2010", "28/10/2010",
"29/08/2010"), class > "factor"),
> end_date = structure(c(18L, 34L, 16L, 12L, 15L, 14L, 29L,
> 41L, 32L, 7L, 19L, 17L, 24L, 36L, 26L, 30L, 36L, 31L, 27L,
> 11L, 36L, 36L, 26L, 28L, 9L, 20L, 41L, 15L, 25L, 24L, 10L,
> 40L, 29L, 5L, 11L, 33L, 4L, 28L, 18L, 29L, 1L, 22L, 14L,
> 1L, 30L, 18L, 24L, 30L, 33L, 30L, 23L, 30L, 30L, 37L, 1L,
> 4L, 30L, 4L, 30L, 33L, 4L, 30L, 33L, 4L, 30L, 3L, 39L, 42L,
> 6L, 38L, 2L, 21L, 8L, 35L, 35L, 13L), .Label =
c("01/02/2012",
> "02/06/2011", "03/02/2011", "03/02/2012",
"03/05/2012", "03/06/2011",
> "04/01/2011", "04/02/2011", "05/03/2011",
"05/06/2012", "07/04/2011",
> "09/04/2011", "10/02/2012", "10/04/2011",
"10/04/2012", "10/07/2012",
> "10/12/2010", "11/04/2011", "12/10/2011",
"13/01/2011", "13/02/2012",
> "13/11/2012", "14/02/2012", "15/02/2012",
"16/01/2012", "17/02/2012",
> "18/05/2012", "18/07/2012", "19/07/2011",
"19/07/2012", "19/10/2011",
> "20/06/2011", "20/07/2011", "21/02/2012",
"21/10/2012", "23/02/2011",
> "23/03/2011", "24/01/2011", "26/01/2011",
"28/01/2011", "29/03/2012",
> "30/05/2011"), class = "factor")), .Names =
c("animal", "collar",
> "start_date", "end_date"), class =
"data.frame", row.names = c(NA,
> -76L))
>
>
> I have so far been able to make some conditional tests, though I would
> need to do this for all 76 animals in the
>
> study...
>
> animal1test<- (date>=01/01/13 && date<=03/01/13)
> animal1test2<- (date>=04/01/13 && date<=06/01/13)
> animal2test<- (date>=04/01/13 && date<=06/01/13)
>
> to use in a giant'if else' formula:
>
> if(animal1test){
> collar[1]="animal1"
> } else if(animal1test2){
> collar[5]="animal1"
> }else if(animal2test)
> collar[1]="animal2"
> }else "NA"
>
> ...etc...
>
> As I'm sure you can see, this is completely inelegant, and also not
> working for me! Any ideas on how to a achieve
>
> this?
>
> Thanks,
> Cat
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
____________________________________________________________
GET FREE SMILEYS FOR YOUR IM & EMAIL - Learn more at
http://www.inbox.com/smileys
Works with AIM?, MSN? Messenger, Yahoo!? Messenger, ICQ?, Google Talk? and most
webmails