[R] Identifying peaks (or offsets) in a time series

[Gabor Grothendieck]

Try:

RSiteSearch("finding peaks")


On 7/24/06, Tauber, Dr E. <et22 at leicester.ac.uk> wrote:
> Dear R-users,
>
> We are monitoring the activity of animals during a few days period. The
> data from each animal (crossing of infra-red beam) are collected as a
> time series (in 30 min bins). An example is attached below.
>
> y <-
> c(0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,3,28,27,46,76,77,60,19,35,55,59,48
> ,87,20,38,82,62,60,85,105,69,109,102,100,101,116,126,119,63,27,25,15,8,0
> ,0,3,0,0,3,0,0,5,3,0,0,6,1,29,73,56,56,57,92,34,51,30,76,30,38,47,87,22,
> 0,68,76,94,101,119,114,115,111,116,134,125,76,23,19,30,2,8,0,3,0,0,0,7,0
> ,0,0,0,4,0,7,0,21,4,49,51,56,43,55,55,34,48,16,0,61,22,94,63,102,47,100,
> 96,113,93,109,123,120,124,115,94,96,76,36,3,0,0,0,0,0,0,2,5,0,0,0,0,2,10
> ,33,34,15,0,47,22,20,33,52,4,41,45,0,21,18,38,32,21,78,82,72,102,103,118
> ,116,118,114,82,18,5,21,4,0,14,0,5,2,0,0,2,2,0,0,3,0,2,7,16,13,17,50,0,4
> 8,16,19,34,39,33,3,67,0,68,34,65,84,61,100,85,108,124,141,139,134,96,54,
> 91,54,12,0,0,0,0,0,0,0,0,0,0,0,4,11,0,19,27,15,12,20)
>
> We would like to have an automatic way, using R, to identify the time
> point of offset of each bout of activity (i.e. when activity goes down
> to a minimum value, for a defined duration). In the example above the
> offset times (the element number) should be approximately: 53, 99, 146,
> 191, 239, 283, 330 (the last bout of activity can be ignores).
>
> Any help or advice will be greatly appreciated,
>
> Many thanks, Eran
>
>
> Eran Tauber (PhD)
> Lecturer in Molecular Evolution
> Dept. of Genetics
> University of Leicester
> University Rd, Leicester LE1 7RH
> England
> ____________________________________________________________
> Phone: 44 (0)116 252-3455, 252-3421 (lab) Fax: 44 (0)116 252-3378
> www.le.ac.uk/genetics/et22
>
> ______________________________________________
> R-help at stat.math.ethz.ch 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.
>