R help - Jan 2011 - Selecting the first occurrence of a value after an occurrence of a different value

Hello everyone, 

I am currently working with some data where I need to select the first
occurrence of a value after the occurrence of another value. 

The data has two columns, one with a time and one with occurence of certain
events. 


The column of data I want to select from looks like this (and each of these
events have a corresponding time in another column). 

Stat71
OutMag
FirstResp
InMag
MagUp 
OutMag
MagDwn
Resp  
Resp
Resp
InMag
MagUp 
OutMag
InMag
OutMag
InMag
OutMag
InMag
OutMag
InMag
MagDwn
OutMag
Resp  
MagUp 
InMag
MagDwn
OutMag
Resp  
MagUp 


Using "which" I have selected all the individual occurrences of each
event
and created variables with the time of said events within them which I used
to calculate differences in time etc for analysis.

But now I am faced with the issue that I need to select only the first
"Resp" that occurs after a "MagDwn" so that I can calculate
the differences
in time between these two events.

Any help would be appreciated, usually I just read these forums and try and
figure out problems for myself but this is for my masters and I'm running
out of time to get it finished :)

Thanks

Surrey 

-- 
View this message in context:
http://r.789695.n4.nabble.com/Selecting-the-first-occurrence-of-a-value-after-an-occurrence-of-a-different-value-tp3217340p3217340.html
Sent from the R help mailing list archive at Nabble.com.

Hi,

 I'll do something like this (there is probably a nicer way, but this code
is what first comes up to my mind)

 provided your data is in x

 pos1 <- which(x %in% val1)[1]
 pos2  <- which(x %in% val2)[which(x %in% val2) >  pos1][1]

 
-- 
View this message in context:
http://r.789695.n4.nabble.com/Selecting-the-first-occurrence-of-a-value-after-an-occurrence-of-a-different-value-tp3217340p3217385.html
Sent from the R help mailing list archive at Nabble.com.

Thanks so much that works, but I just realised I was unclear, I meant I need
to select all the "MagDwn" after every "Resp" :)
-- 
View this message in context:
http://r.789695.n4.nabble.com/Selecting-the-first-occurrence-of-a-value-after-an-occurrence-of-a-different-value-tp3217340p3217456.html
Sent from the R help mailing list archive at Nabble.com.

Ofcourse you can loop over your data, but a vectorised way:


f.search <- function(x, ref, search) {
  cs <- match(cumsum(x==ref), cumsum(x==ref))
  outp <- suppressWarnings(tapply(x==search,cs, function(x)
min(which(x==1))))
  outp <- outp[outp!=Inf] # To remove the occurences where nothing was found
  return(as.vector(as.numeric(names(outp))+outp - 1))
}

f.search(x, "MagDwn", "Resp")

Good luck

Bart
-- 
View this message in context:
http://r.789695.n4.nabble.com/Selecting-the-first-occurrence-of-a-value-after-an-occurrence-of-a-different-value-tp3217340p3217482.html
Sent from the R help mailing list archive at Nabble.com.

Try this:

 which(x %in% c("MagDwn", "Resp"))[2]

On Fri, Jan 14, 2011 at 7:22 AM, surreyj <surreyjackson@gmail.com> wrote:
>
> Hello everyone,
>
> I am currently working with some data where I need to select the first
> occurrence of a value after the occurrence of another value.
>
> The data has two columns, one with a time and one with occurence of certain
> events.
>
>
> The column of data I want to select from looks like this (and each of these
> events have a corresponding time in another column).
>
> Stat71
> OutMag
> FirstResp
> InMag
> MagUp
> OutMag
> MagDwn
> Resp
> Resp
> Resp
> InMag
> MagUp
> OutMag
> InMag
> OutMag
> InMag
> OutMag
> InMag
> OutMag
> InMag
> MagDwn
> OutMag
> Resp
> MagUp
> InMag
> MagDwn
> OutMag
> Resp
> MagUp
>
>
> Using "which" I have selected all the individual occurrences of
each event
> and created variables with the time of said events within them which I used
> to calculate differences in time etc for analysis.
>
> But now I am faced with the issue that I need to select only the first
> "Resp" that occurs after a "MagDwn" so that I can
calculate the differences
> in time between these two events.
>
> Any help would be appreciated, usually I just read these forums and try and
> figure out problems for myself but this is for my masters and I'm
running
> out of time to get it finished :)
>
> Thanks
>
> Surrey
>
> --
> View this message in context:
>
http://r.789695.n4.nabble.com/Selecting-the-first-occurrence-of-a-value-after-an-occurrence-of-a-different-value-tp3217340p3217340.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help@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.
>


-- 
Henrique Dallazuanna
Curitiba-Paraná-Brasil
25° 25' 40" S 49° 16' 22" O

	[[alternative HTML version deleted]]

Thanks so much for your help everyone, got it sorted now :)

this is what I used in the end:

timeofonlyfirstresponseafterrft<-testdata$Time[test<-which(!diff(as.numeric(factor(Stat,
levels = c("MagDwn", "Resp")))))]
timeofonlyfirstresponseafterrft1<-as.POSIXct(timeofonlyfirstresponseafterrft,
origin="timeofstart[1]", format="%Y,%m,%d, %H, %M, %OS")

Cheers

Surrey
-- 
View this message in context:
http://r.789695.n4.nabble.com/Selecting-the-first-occurrence-of-a-value-after-an-occurrence-of-a-different-value-tp3217340p3218653.html
Sent from the R help mailing list archive at Nabble.com.

Hello, 

Back again, 

I thought the problem was solved but I realised that the only reason I was
getting the correct answer was because my data set happened to only have two
"rfts" to choose from, so it looked correct.

I have been using:

onlyfirstresponseafterrft<-which(!diff(as.numeric(factor(Stat, levels
c("MagDwn", "Resp")))))
> onlyfirstresponseafterrft
  [1]  11  12  13  19  20  21  27  28  29  35  36  37  43  44  45  51  52 
53
 [19]  59  60  61  67  68  69  75  76  77  83  84  85  91  92  93 103 104
105
 [37] 113 114 115 121 122 123 129 130 131 137 138 139 145 146 147 153 154
155
 [55] 161 162 163 173 174 175 181 182 183 191 192 193 199 200 201 207 208
209
 [73] 215 216 217 225 226 227 233 234 235 241 242 243 251 252 253 259 260
261
 [91] 267 268 269 275 276 277 283 284 285 291 292 293 303 304 305 311 312
313
[109] 319 320 321 329 330 331 337 338 339

to get my results and what is being delivered is the rows at which a
"resp"
occurs after a "magdwn" except I only want the first "resp"
after a mag
down... This seems simple to figure out and I have tried a lot of things but
its just not happening!

Can anyone help me please?

Thanks
-- 
View this message in context:
http://r.789695.n4.nabble.com/Selecting-the-first-occurrence-of-a-value-after-an-occurrence-of-a-different-value-tp3217340p3220788.html
Sent from the R help mailing list archive at Nabble.com.

Hi Freddy, 

I have a long column of event codes e.g. below (in multiple files) that I am
trying to analyse.  

OutMag
FirstResp
InMag
MagUp
OutMag
MagDwn
Resp
Resp
Resp
InMag
MagUp
OutMag
InMag
OutMag
InMag
OutMag
InMag
OutMag
InMag
MagDwn
OutMag
Resp
MagUp
InMag
MagDwn
OutMag
Resp
MagUp

With these files I have been using which to select the appropriate event
codes so I can do analysis on timing between events using the time
appropriate time coloumn. 

This has worked well so far and now I am faced with the problem that I need
to select the first "Resp" that occurs after each "MagDwn". 
Sometimes there
will be just one "Resp" in between a "MagDwn" and sometimes
there will be
many e.g. up to 500.  

This code onlyfirstresponseafterrft<-which(!diff(as.numeric(factor(Stat,
levels = c("MagDwn", "Resp")))))
allowed me to select the first "Resp" (or so I thought) but in the
file I
tried it on there were only two occurences of "Resp" inbetween each
"MagDwn"
and now that I have tried it on files with more "Resp" it is actually
selecting all but the last one that happens before the "Mag Dwn" but I
only
need the first one. 

Hope that makes sense, thanks for your help. 
-- 
View this message in context:
http://r.789695.n4.nabble.com/Selecting-the-first-occurrence-of-a-value-after-an-occurrence-of-a-different-value-tp3217340p3220852.html
Sent from the R help mailing list archive at Nabble.com.

On Sun, Jan 16, 2011 at 11:09:58PM -0800, surreyj wrote:
> 
> Hello, 
> 
> Back again, 
> 
> I thought the problem was solved but I realised that the only reason I was
> getting the correct answer was because my data set happened to only have
two
> "rfts" to choose from, so it looked correct.
> 
> I have been using:
> 
> onlyfirstresponseafterrft<-which(!diff(as.numeric(factor(Stat, levels
> c("MagDwn", "Resp")))))
[...]
> 
> to get my results and what is being delivered is the rows at which a
"resp"
> occurs after a "magdwn" except I only want the first
"resp" after a mag
> down... This seems simple to figure out and I have tried a lot of things
but
> its just not happening!
Is it required to select positions with Resp, which are the
end-points of subsequences of the form MagDwn other^* Resp ?
For the sequence

  1  MagDwn
  2  other 
  3  MagDwn
  4  Resp  
  5  other 
  6  Resp  
  7  MagDwn
  8  other 
  9  Resp  

this would be positions 4 and 9.

The positions of Resp in these end-points may be computed,
for example

  Vals <- c("MagDwn", "Resp", "other")
  Stat <- Vals[c(1, 3, 1, 2, 3, 2, 1, 3, 2)]
  ind <- which(Stat %in% c("MagDwn", "Resp"))
  Reduced <- Stat[ind]
  ind[which(diff(Reduced == "Resp") == 1) + 1]
  # [1] 4 9

The positions of the corresponding MagDwn are

  ind[which(diff(Reduced == "Resp") == 1)]
  # [1] 3 7

Petr Savicky.

Thanks so Peter, works great!

Surrey 
-- 
View this message in context:
http://r.789695.n4.nabble.com/Selecting-the-first-occurrence-of-a-value-after-an-occurrence-of-a-different-value-tp3217340p3221271.html
Sent from the R help mailing list archive at Nabble.com.