Sure, but it seems like the point of the time series class is so you don't have to create a factor based on the period of sampling. Is there anyway of imposing calculations like median on the time series class, or is Kjetil's suggestion the only approach?
Thanks! On Sun, Nov 1, 2009 at 4:09 PM, Kjetil Halvorsen < [email protected]> wrote: > introduce a factor variable with the months and then use tapply? > > Kjetil > > On Sun, Nov 1, 2009 at 9:07 PM, Tim Bean <[email protected]> wrote: > > Hello, I have a quick question about time series methodology. If I want > to > > display a boxplot of time series data, sorted by period, I can type: > > > > boxplot(data ~ cycle(data)); > > > > where data is of class "ts" > > > > Is there a similar method for calculating, say, the median value of each > > time step within the series? (So for a monthly data set, calculate median > > for all Januarys, all Februarys, all Marchs, etc.) > > > > Thanks, > > Tim > > > > [[alternative HTML version deleted]] > > > > ______________________________________________ > > [email protected] 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. > > > [[alternative HTML version deleted]] ______________________________________________ [email protected] 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.
