|
PureBytes Links
Trading Reference Links
|
Dear Mr. John Lynch:
Thank you for attempting to reproduce the CSI data problems.
Even if you are not able to reproduce the problem right now,
please do not assume that CSI Tech Support that CSI is
"having difficulty reproducing Leslie's problems".
Tech Support is aware of the problem and is trying to help.
The issue went all the way to the President of CSI,
Mr. Bob Pelletier.
Here is part of Email from Mr. Bob Pelletier dated Friday,
Oct 26:
Pelletier write to me:
Admittedly something was wrong then and
we will attempt to resolve the anomaly.
What is happening is that for some unknown
reason whenever the reported open interest
goes over 900,000 in total, the CBOT reports
a value net of 900,000. This error has caused
quite a bit of trouble for some customers
including us and you.
{End of quote form Pelletier}
{A value net of 900,000 is one order of magnitude error.
Enough to blow up most trading systems. }
So, the good news is that Pelletier (and his company, CSI)
are indeed responsive, recognize the problem, and attempting to
make corrections.
CSI has corrected the negative volume issue in the NG contract.
>From an Oct 26th perspective, the August data looks clean.
The reason you did not see the error is that CSI sends out
error corrections. (This can be next day, next week or any time.)
Unfortunately, from an August of 2002 perspective,
the NG volume was in error. The error was not corrected
until mid September.
Since we all trade the data that comes in, we all suffer when the
data is wrong. We cannot send error corrections to our broker
and ask for our money back.
Hopefully, with Pelletier's involvement, CSI or CBOT, or
whomever will find the "unknown reason" and fix the error.
For right now, anyone who uses OI, Volume, in trading or in
rolling their contracts needs to be vigilant and inspect
the data every day that it comes in.
Respectfully yours,
Leslie Walko
John Lynch wrote:
>
> I have found CSI to be very responsive and keen to help with any problems so
> I thought that I'd drop a copy of Leslies email to support@xxxxxxxxxxx . I
> got a response this morning seeking my view of the problems. It seems that
> they are having difficulty reproducing Leslies problems and asked me if I
> got the same issues.
>
> I checked my data and found the results below. Basically I can't see
> the problems Leslie is alerting us to but if anyone else does I am sure that
> csi would appreciate you letting them know including your version of UA in
> the email.
>
> John
>
> "not affiliated in any way with any vendor - just another customer :-)"
>
> MY RESULTS (as forwarded to CSI):
>
> I dont have the negative numbers either but the volume for TY looked a bit
> light on the dates he specified. I dont care about volume really but
> thought I should alert you to the possible problem. Lets see what I have:
>
> I'm running 2.4.4 which I am happy with so I havent moved to 2.5.
> For my backadjusted contracts I adjust on 2nd consecutive day of higher OI.
> I am aware that the last day is always a fiction and in the contracts below
> gives a very low number.
>
> So heres a go:
>
> 1) TY20_IOB 20021021 vol 84536 looks light.
> 2) 20021011 looks light but not stunningly at 95305
> 2b) I cant understand Leslies issue with the 14th of October ... I have no
> trading on that day and I seem to recall that it was a holiday for certain
> markets including the financials (Columbus??).
>
> but ... all of the OI figures look ok.
>
> 3) EC2 and JY2 OI figures look just fine.
>
> 4) NG I dont have any negative numbers but perhaps Leslie had them
> historically?
>
> 5) Soy beans and TY missing prices - between Sept 5 to 10th. I didnt have
> any problems and I look at them every day so I would notice.
>
> Regards, John
>
> ----- Original Message -----
> From: "Leslie Walko" <l.walko@xxxxxxxxxxx>
> To: "Omega List" <omega-list@xxxxxxxxxx>
> Sent: Thursday, October 24, 2002 11:30 AM
> Subject: Waring: CSI data errors
>
> Anyone trading CSI data should be on the lookout for
> Volume and Open Interest errors, missing prices, and
> NEGATIVE volume.
> I found all of these errors in their contract since
> August. See list below.
>
> This may effect your system even if your analysis ignores
> volume. Why? Many traders roll contracts on OI,
> or OI+Volume, thus errors in volume causes strange rolling
> effects. (Remember, once your rolled, you cannot UNROLL,
> and go back to the prior contract.)
>
> Ten Year Note:
> Right now, their Ten Year note, day session, TY2
> Contract Volume figures are dropping the leading character.
> The TY2 Open Interest is off by 900,000 contracts.
> (Yes, Nine Hundred THOUSAND.)
> Look at dates for today, yesterday, Oct. 11, Oct. 14.
>
> Euro Currency, Japanese Yen:
> I just had a quick look.
> Volume and OI are missing digits.
> (An order of magnitude error!)
>
> Natural Gas:
> In August, CSI was reporting NEGATIVE volume on NG,
> Natural Gas contract on my birthday (Aug 24th) and
> for a week prior to and subsequent.
>
> (I felt really special when I took a quick 20,000 drawdown
> due to their data errors.)
>
> Soy Beans and Ten Year:
> Between Sept 5 to 10th, they failed to report prices for
> some commodities icluding TY2 and Soy Beans.
>
> If you note errors, please call them to report the error,
> and impress upon them the importance of accuracy.
>
> Leslie
>
> --
> Regards,
> Leslie Walko
> 610-688-2442
> --
> "Life is a tragedy for those who feel, a comedy for those who
> think"
> Horace Walpole, 4th earl of Orford, in a letter dated about 1770
--
Regards,
Leslie Walko
610-688-2442
--
"Life is a tragedy for those who feel, a comedy for those who
think"
Horace Walpole, 4th earl of Orford, in a letter dated about 1770
|