|
PureBytes Links
Trading Reference Links
|
Nightmare.
Some of those text filings are megabytes in length. Rarely are they
the same format, even for the same company from year to year.
Different industries have significantly different formats.
The various online services (MSN, Yahoo, AdvFN, etc) can't even agree
how to categorize the data that is already there. One will put a line
item in one of their pigeonholes, another will put it into a different
one. I've even seen differences in something as "simple" as "Cash &
Equivalents" (e.g. "Restricted Cash"). Although there are a many
similarities between statements, how are you going to automate all of
the "one time" special line items that are unique for an individual
company and an individual filing?
AdvFN is my favorite source, since it has the most history. However,
it has been noted to me that MSN has different data from AdvFN because
they "fill in" missing quarters with information from press releases.
Information that hasn't been filed with the SEC because it wasn't
required. And, it appears AdvFN may not pick up restatements of SEC
filings.
On 2/9/07, eodeen <eodeen@xxxxxxxxx> wrote:
>
> After reading the threads below about fundamental database &
> backtesting in AB I decided to share the following idea. Why not
> retrieve the fundamental data directly from the official 10-Q forms
> filed at edgar?
> This data is FREE including historical 10-Q forms. Take IBM for
> example, one can retrieve the 10-Q forms in xml or text format back to
> 1994.
Content-Description: "AVG certification"
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.5.441 / Virus Database: 268.17.34/679 - Release Date: 2/10/2007 4:08 PM
|