[R] Aggregating frequency of irregular time series

[Ivan Alves]

Thank you very much Brian.  Indeed, by looking at ?tapply() this would 
do the job for regular time series (whose data-time INDEX renders 
itself naturally for " bundling data" in regular groups). However, with 
irregular time series the story is different, as some careful 
"bundling" is necessary prior to applying tapply(). Whereas this may 
not be a problem for months (creating strings month-year), it would be 
for weeks, for instance.  Also the object returned would need to be 
converted again to a time series object with additional function calls. 
Furthermore, the function AggregateSeries() provides additional 
functionality, such as creating moving averages, rolling variances, 
minima and maxima, all with options to the same function call. I take 
from your response that there is no easy way out (an already existing 
function), and that some programming will be required.

Kind regards,

Ivan
_______________________
Ivan Alves
mailto://papucho@mac.com
On 25 Mar 2004, at 22:51, Prof Brian Ripley wrote:

> R itself has no support for irregular time series, but it does have an
> aggregate method for regular ones.  You need to look into whichever
> package is handling irregular time series.  However, it seems to me 
> that
> this is not a time series problem at all: you have a set of 
> observations
> whose indices are data-times, and tapply() will do the job.
>
> On Thu, 25 Mar 2004, Ivan Alves wrote:
>
>> S-Plus has the function AggregateSeries() whose name is self
>> explanatory.  For instance one can derive monthly series from daily
>> ones by specifying end-of-period, averages, sums, etc.  I looked for a
>> similar function in the packages "its" and "tseries", but found
>> nothing.  I also help.searched() for aggregate to no avail.  Would
>> anybody be so kind to point me in the right direction?
>
> -- 
> Brian D. Ripley,                  ripley at stats.ox.ac.uk
> Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
> University of Oxford,             Tel:  +44 1865 272861 (self)
> 1 South Parks Road,                     +44 1865 272866 (PA)
> Oxford OX1 3TG, UK                Fax:  +44 1865 272595
>