[R-SIG-Finance] R-SIG-Finance Digest, Vol 38, Issue 4
a.trapletti at swissonline.ch
Fri Jul 6 12:47:17 CEST 2007
Just my 2 cents: It might be more useful to split the data part from the
analysis or trading or whatever part. In the first step, data is
received and stored in a database. Typically a C/C++ or Java application
receives data and saves it in a database. In a second step,
applications, e.g., R, may access the data through the database, either
access historical data or data in realtime. If well designed, the delay
of receiving data in realtime through the database instead of a direct
connection is negligible. And you gain a lot of flexibility with this
>Date: Thu, 5 Jul 2007 10:54:02 -0500
>From: Dirk Eddelbuettel <edd at debian.org>
>Subject: Re: [R-SIG-Finance] [SPAM] - Re: Intraday data with
> RBloomberg - Email found in subject
>To: <davidr at rhotrading.com>
>Cc: r-sig-finance at stat.math.ethz.ch, Ian Seow <ianseow at gmail.com>
>Message-ID: <18061.5146.455992.808579 at basebud.nulle.part>
>Content-Type: text/plain; charset=us-ascii
>On 5 July 2007 at 09:56, davidr at rhotrading.com wrote:
>| I can verify similar behavior.
>| I can get the data via VBA behind Excel.
>| Between versions of Bloomberg and R and the difficulty I seem to have
>| getting RDCOMClient installed correctly, I have sort of given up on this
>| approach, unfortunately so, since it seems so useful. I pretty much use
>| VBA/Excel or C# to generate text files to read into R.
>Well, the C/C++ code I wrote two employers ago, and which is hence still
>owned by that firm and hence no shareable, worked pretty well.
>It shouldn't take too long to rewrite this, starting from the Bloomberg C API
>examples. As I don't currently use or need Bloomberg, I can't help. Maybe
>somebody else can though.
Phone : +41 (0) 44 9945630
Mobile : +41 (0) 76 3705631
Email : a.trapletti at swissonline.ch
More information about the R-SIG-Finance