[R] Interpolate irregular time series

[Gabor Grothendieck]

On Fri, Oct 22, 2010 at 9:20 AM, Alexander Salim <salim at gmx.ch> wrote:
> Hi all,
>
>
>
> Issue:
>
> I have two datasets, one is a regular time series (rain gauge) with
> resolution of 10 minutes. The other one is an irregular time series (link).
> Now I want to analyze the correlation between these two datasets with linear
> regression. The regular time series is a data.frame and the irregular time
> series is a zoo object.
>
>
>
> Problem:
>
> For the irregular time series (link) I want also a 10 minutes resolution.
> Therefore I smoothed the curve and with the function 'loess.smooth()' it's
> possible to evaluate the points of the curve. R has a number code for every
> date (e.g. "2009-05-01 00:02:11" = 1241136131). My idea is to use this
> number code to receive the function value of the smoothed curve. So far,
> unsuccessfully.
>
>
>
> I read a lot of manuals but I couldn't find an answer. If this issue is
> already answered I apologize.
>

If the problem is to align two series then see #13 in the zoo FAQ.
  vignette("zoo-faq")

-- 
Statistics & Software Consulting
GKX Group, GKX Associates Inc.
tel: 1-877-GKX-GROUP
email: ggrothendieck at gmail.com