[R] moving average on irregular time series

[Gabor Grothendieck]

Replace the non-events with NA and then use na.locf from the zoo
package to move the last event date up to give lastEvent.
Then simply select those rows whose lastEvent date is at least 14 days
ago or if the row itself is an Event:

> library(zoo) # na.locf
>
> lastEvent <- with(exData, na.locf(ifelse(Event, Datebegin, NA), na.rm = FALSE))
> exData[beg >= lastEvent + 14 | exData$Event, ]
    Datebegin Event
1  2009-08-20  TRUE
2  2009-11-21 FALSE
3  2009-11-28 FALSE
4  2010-01-12 FALSE
5  2010-01-18 FALSE
6  2010-01-23 FALSE
7  2010-02-21 FALSE
8  2010-03-03 FALSE
9  2010-04-06 FALSE
10 2010-04-16 FALSE
11 2010-05-22  TRUE
12 2010-05-27  TRUE
13 2009-09-08  TRUE
14 2009-10-01 FALSE
15 2009-10-06 FALSE
16 2009-10-09 FALSE
17 2009-10-11  TRUE
19 2009-10-28 FALSE
20 2009-10-31 FALSE


On Thu, Jun 3, 2010 at 10:23 AM, Gustaf Rydevik
<gustaf.rydevik at gmail.com> wrote:
> Hi all,
>
>
> I wonder if there is any way to calculate a moving average on an
> irregular time series, or use the rollapply function in zoo?
> I have a set of dates where I want to check if there has been an event
> 14 days prior to each time point in order to mark these timepoints for
> removal, and can't figure out a good way to do it.
>
> Many thanks in advance!
>
> Gustaf
>
>
> Example data:
>
> exData<-structure(list(Datebegin = structure(c(14476, 14569, 14576, 14621,
> 14627, 14632, 14661, 14671, 14705, 14715, 14751, 14756, 14495,
> 14518, 14523, 14526, 14528, 14529, 14545, 14548), class = "Date"),
>    Event = c(TRUE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE,
>    FALSE, FALSE, FALSE, TRUE, TRUE, TRUE, FALSE, FALSE, FALSE,
>    TRUE, FALSE, FALSE, FALSE)), .Names = c("Datebegin", "Event"
> ), row.names = c(NA, 20L), class = "data.frame")
>
> ###In this example, row 18 is a date less than 14 days after an event
> and should be marked for removal.
>
>
>
> --
> Gustaf Rydevik, M.Sci.
> tel: +46(0)703 051 451
> address:Essingetorget 40,112 66 Stockholm, SE
> skype:gustaf_rydevik
>
> ______________________________________________
> 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.
>