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

Applied Derrivatives Trading site and TS critique


  • To: omega-list@xxxxxxxxxx
  • Subject: Applied Derrivatives Trading site and TS critique
  • From: "Bill M." <billmasi@xxxxxxxx>
  • Date: Mon, 3 May 1999 21:48:25 -0400 (EDT)
  • In-reply-to: <199905031659.JAA30332@xxxxxxxxxxxxxx>

PureBytes Links

Trading Reference Links

   
I don't know if the list is aware of the European site called "Applied
Derivatives Trading".  It's first-rate.  Good material published every month.

This month features the following about Omega's latest.

URL of latest issue is http://www.adtrading.com/adt38

Bill M.




May '99 Feature:
Tradestation Miscalculates?  [Title]


We recently received the following interesting e-mail from Sean Boughal and
thought it deserved to be featured to bring it to readers' attention: 

Tradestation 2000i Miscalculation Of Upticks/Downticks


Dear ADT, 

Congratulations on your 3rd B-Day!! You guys are the Avengers of the
Financial Universe!! 

I don't know whether you guys care, but if you use Tradestation you might.
Considering your clout in the industry, I figured I'd let you know of a
potential problem in the new version of Tradestation (2000i). I just
upgraded and, in the process of upgrading my analysis techniques, found out
that the Reserved Words Upticks and Downticks are calculating incorrectly! 

I have compared TS 4.0 (old version) to TS 2000i (new version) on the same
data and even gone over it manually (counting each tick). TS 4.0 calculates
correctly, while TS 2000i over counts Upticks and undercounts Downticks.
For me, and many other traders, this is very unfortunate because my
analysis relies heavily on Upticks vs. Downticks. I have tried informing
Omega Research of the flaw in the way TS 2000i calculates
Upticks/Downticks. Unfortunately, they fail to recognize the problem. The
e-mail responses I have gotten from their Tech. support have been laughable
at best! In their explanations to me they have even gone as far as making
up a Reserved Word (UnchangedTicks). 

The following is a description of how Upticks and Downticks is calculated
in the new TS 2000i vs. TS 4.0 according to my tests: 

TS 2000i: 

Upticks = Any Tick that is Greater or Equal to the previous Tick,
regardless of the most recent tick change direction. 

Downticks = Any Tick that is Less than the previous Tick. 

vs. 

TS 4.0: 

Upticks = Any Tick that is Greater, or Equal to the previous Tick, as long
as the last change in tick direction was positive. 

Downticks = Any Tick that is Less than, or Equal to the previous tick, as
long as the last change in tick direction was negative. 

As you can see, TS 2000i over counts Upticks and under counts Downticks
which results in faulty analysis of Market conditions (e.g. Upticks >
Downticks in a down trending price cycle). In order to accurately compare
Upticks to Downticks (supply and demand), they must be contrary to each
other in the means that they are calculated. This can be done by either
counting only Upticks and Downticks (thus taking EqualTicks into account as
neutral), or by using the old TS 4.0 means. 

My tests show that on historical data TS 2000i has 2 x more Upticks vs
Downticks than TS 4.0! (TS 4.0 has close to an equal distribution of Up vs.
Down ticks). A perfect Market, Bull phase followed by a Bear Phase of
opposite Amplitude and Frequency would have an Uptick/Downtick distribution
exactly equal to another (#Upticks = #Downticks). 

I believe they attempted to give a neutral weighting to Equal Ticks,
neither counting as Up or Down Tick, but got lost along the way while
programming. Even if they meant to program it this new way, Buying vs
Selling pressure can not accurately be analyzed using Upticks - Downticks
and an important tool is lost. I know many traders who refuse to upgrade to
TS 2000i because their analysis also relies heavily on Upticks vs. Downticks. 

I just figured I'd let you know, hopefully I made some sense to you. Keep
up the good work and thanks again! 

Sincerely, 


Sean Boughal 



----------------------------------------------------------------------------
----


Ed: Any other foibles readers? All correspondence will be received with
interest.