[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: TS 2000 History Bank: ABSOLUTELY useless for 1 minute data



PureBytes Links

Trading Reference Links

Chris:
I agree with you completely; Omega should provide users with the option to
download daily, minute and/or tick data.  It should also be easy to download
data for a simple list of symbols rather that a whole exchange worth.  But
most importantly the data should be accurate which is not the case
currently.
There is another reason why using tick data from HBank to build minute data
is unwise: HBank's tick data in my experience is nearly worthless due to
missing ticks.  On the several occasions when I have downloaded tick data in
an effort to recover data after Global Server crashes, I discovered that
most of the day's ticks were missing.  Some stocks only had about 1/3 of the
total ticks actually traded that day.  Accordingly minute bars constructed
from such data would be near worthless and misleading.  If you are a stock
trader and follow intraday volume missing ticks are disasterous since volume
is understated by the size of every missing tick.
Bill

> -----Original Message-----
> From: Chris Baker [mailto:chrisbak@xxxxxxxxx]
> Sent: Sunday, August 15, 1999 9:35 AM
> To: Omega List
> Subject: Re: TS 2000 History Bank: ABSOLUTELY useless for 1 minute data
>
>
> All I want is 1 minute data, not Tick Data.   That should be a very fast
> download, even for 40-50 stocks, for MONTHS of data.    However
> my numerous
> attempts to download even Tick Data for 40-50 stocks on a single day, such
> as August 2, failed.
>
> I would think it would be easiest for users if Omega to provided
> one minute
> data for a fast download of MONTHS of data at a time.
> Downloading unwanted
> Tick Data could take 100 times as long for big-cap stocks, if it worked.
>
> I would think providing one minute data would have been an obvious design
> objective for History Bank, especially when Omega advertises they are
> selling the best trading application for back-testing.
>
>
>
>