Message-ID: <8ec76080911190652l2d7a69b7wf3c3e37ed8697c5d@mail.gmail.com>
Date: 2009-11-19T14:52:30Z
From: Whit Armstrong
Subject: Add values to time series in DB directly
In-Reply-To: <E5BB21E7-F714-4FB0-B5CF-F2EAFD1D838A@bgki.net>
just append the new data to the table. as long as you have a way of
subsetting just the new rows, this should be easy. I haven't used
TSdbi, but I have a postgres package that uses binary write, so it's
pretty fast. http://github.com/armstrtw/unifieddbi
-Whit
On Thu, Nov 19, 2009 at 9:32 AM, Josuah Rechtsteiner
<rechtsteiner at bgki.net> wrote:
> Dear members of the mighty list,
>
> I have a PostgreSQL DB as storage for some time series. I access the DB from
> within R with TSPostgresql/TSdbi. Today I asked myself if it is possible to
> update an existing TS in the DB with new data points without having to load
> the whole time series into R first and then replacing it in the DB.
> I tried
>
> TSreplace(x, con, append=TRUE)
>
> but this did not work. After reading the vignettes and manuals of the
> packages in use here without success, I thought I should probably conjure up
> the mighty list...
>
> Best,
>
> Josuah
>
> _______________________________________________
> R-SIG-Finance at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only.
> -- If you want to post, subscribe first.
>