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

bouncing ticks - the BIG COMBO answer



PureBytes Links

Trading Reference Links


Ok boys here is the BIG COMBO answer to everyone

Canyon678@xxxxxxx wrote:
> Hi Hans, Editorial,Mark, 
> 
> Good notes like this die down . I dont know why ?

for me: Im flooded with private messages, suggestions, ideas, workarounds 
(tm OR) ALL good and bad and testing those to overcome the problem:

I **think** using 405 minute bar charts may be the best for S&P - Im not sure, 
as Im still stunned by that discovery I made here REALTIME !

TJ wrote:
> now that ya mentioned it, i'd like to know why mr zebra first said that "+
> I know of BOUNCING TICKS - as some people tried to explain this problem 
> to me privately - it OBVIOUS cant be part here" and then the "z" said "What
> it is here, is the good ole BOUNCING TICKS - but much worse than I ever
> assumed."
> 
> well, a bouncing tick problem would have been my first troubleshooting
> thingy to suggest, but zebra said otherwise. 

you see - even ZEBRAS can be wrong !

> it suggests that mr. zebra doesn't always know what it's it's talking about, 

glad you are

> or mr. z can not communicate clearly. 

thats what my wife says too.


With "knowing" of the BT_problem for some 4 years, I was sure it cant be part 
here as Im talking REALTIME data received REALTIME and the BT is a 
SIMULATION which OR uses on HISTORICAL data!

in very short: BT is the problem that TS does not know whether the low or high 
was seen first in a DAILY BAR..... agreed on HISTORIC files I can 
understand the problem. BUT my thing has happened REALTIME. Therefore 
I didnt believe it was it - but as people described privately as they had it hapen 
to them too, it really is - therefore I had to change my statement, at least I can 
DO THAT if I see that I been wrong !

>  if the zebra has a white box system, then mr. zebra can code around the
> deficiencies.

for the umpteeth time - yes it can be done - BUT using the BUILT IN STOP is 
a desaster and TS changes the TRADES which have APPEARED on the 
screen in REALTIME after the day with HINDSIGHT - this should not happen 
at all !!!! THATS MY POINT !

in short words: TS IS CHEATING !!!! - to be friendly !

Patrick White wrote:
> With features like this, who needs bugs?

no comment :-)

"Sentinel" <rjbiii@xxxxxxxxx> wrote:
> I have to take a stab at this because some time ago I said on this list
> that my systems were running a little differently on 2000i, this is
> problem that I was referring to.
> 
> When running a system on Daily Bars During the market day TS knows how 
> the Daily Bar is being built and where and when prices on the bar were
> reached, and the system reacts properly. When you close the chart or
> reload the Data it no longer knows this information and reverts to the Old
> Bouncing Tick Theory, 

again: what has happened to me was ALL REAL TIME - no reload or anything 
like that - thats my very point !

> and Omega's assumption of how the Bar was
> formed, and will show different results. I also think that if you generate
> orders for the next bar you would not get your exit until the next day.
> 
> If you were to rewrite the system to refer to Daily Data and run on
> Interaday Data this problem is minimized. 

sure there are WORKAROUNDS (tm OR) - but it just cant be accepted that 
TS changes a KNOWN TRADE with HINDSIGHT knowledge with is clearly 
2nd class to REALTIME knowledge


OurMate <g> wrote:
> So what you are saying, Hans, is that you and Mark Down are WRONG and 
> that PO and I are RIGHT?  What you are saying is that this is indeed a 
> FEATURE and NOT A BUG?  Is that right?

Im glad that PO didnt comment - thats to his credit as he does not have 
realtime data and would not qualify, well done pete. 

To OM: Im positively suprised you havent replied for 24hours - looks as you 
first time had some insight in a problem - in difference to your usual rambling 
of : if you dont PRODUCE you cant complain - thanks


rgds hans

PS: to lots more private mails, which I didnt mention and havent been able to 
answer ....

PSPS: below is the lengthy post from BF which explains the BT problem - I 
still dont accept the fact that this is a VALID excuse here to be used for 
"market simulation" as my data was REALTIME and the trade was REALTIME



> Subj:  CL_Re: Bouncing-Ticks
>  From:    bfulks@xxxxxxxxxxxx (Bob Fulks)
>  At 11:16 AM -0400 6/13/99, Mben@xxxxxxx wrote:
>  
>  >Can anyone explain how bouncing ticks works ? Anyone have any idea as
>  >to how close this comes to replicating actual trading on daily data ?
>  >What assumptions does TS make about order execution on daily bars when
>  >more than one order is triggered ? Is this information documented
>  >anywhere ? I checked on-line help and the manual but could not find
>  >anything. Does anyone know if Bill Brower ever wrote this up in TS
>  >Express ?
>  
>  
>  There is a fairly complete description in the on-line help system in
>  TradeStation 4.0 the text of which is reproduced below. Please see the
>  actual help data for several pictures.
>  
>  Omega often gets criticism for this and you might not agree with using
>  the technique. But they clearly admit that it is a "simulation of
>  market activity" to get around the problem of not having detailed price
>  movement within a bar. And as a "simulation", it might be as good as
>  you can get without more detailed information...
>  
>  Bob Fulks
>  
>  -----
>  
>  Understanding How TradeStation Simulates Market Activity with
>  Historical Data
>  
>  When you apply a trading system to a chart of time-based bars based on
>  historical data, TradeStation uses a set of rules to simulate market
>  activity. It is important to understand how TradeStation handles the
>  simulation of market activity on time-based bars when you are testing a
>  trading system on historical data, for example 60-minute bars or daily
>  bars, based on historical data (non-real-time/delayed data, either
>  intra-day or daily).
>  
>  It is equally important to understand this simulation if you use
>  historical data on time-based bars when you make the trades signaled by
>  a trading system because if you trade a system on time-based bars based
>  on historical data, TradeStation uses the same simulation described in
>  this section.
>  
>  Note: Time-based bars are any bars not based on ticks. For example,
>  1-minute bars, 60-minute bars, daily bars, etc. are all time-based
>  bars. They plot the open, high, low and close price of all transactions
>  that occur within the specified time period. On the other hand, 1-tick
>  bars, 10-tick bars and 50-tick bars are all examples of bars that are
>  based on ticks, or transactions, rather than on time. Tick bars (except
>  for 1-tick bars) plot the open, high, low and close price for the
>  specified number of transactions; 1-tick bars plot the price of a
>  single transaction.
>  
>  The information in the next two headings can help you understand how
>  TradeStation handles the simulation of market activity on time-based
>  bars. The information in the first heading provides a frame of
>  reference by describing how TradeStation handles the entry and exit
>  orders of a trading system applied to time-based bars based on
>  real-time/delayed data.
>  
>  The information in the second section describes the different way
>  TradeStation handles the entry and exit orders of a trading system
>  applied to time-based bars based on historical data (either intra-day
>  data pasted into the Omega Server or daily data in a supported format).
>  Understanding How TradeStation Handles Entry and Exit Orders on
>  Time-Based Bars Built with Real-Time/Delayed Data
>  
>  Every trading system should be designed with two signals: a signal that
>  specifies the conditions in which you want to enter the market, and a
>  signal that specifies the conditions in which you want to exit the
>  market. Generally speaking, a part of each signal is the price level at
>  which you want to enter or exit the market.
>  
>  When you are working with time-based bars based on real-time/delayed
>  data, TradeStation receives price data tick-by-tick (transaction by
>  transaction) and builds time-based bars from that tick-by-tick data. As
>  each transaction is received, TradeStation calculates whether or not
>  the entry or exit criteria of an applied trading system is triggered by
>  that transaction. If a transaction is received that matches the entry
>  criteria of the system, TradeStation issues the entry signal at the
>  exact moment the transaction is received. If a transaction is received
>  that matches the exit criteria of the system, TradeStation issues the
>  exit signal at the exact moment that transaction is received.
>  
>  Therefore, it is possible to both enter and exit the market on the same
>  time-based bar in the event that, within the same time-based bar, a
>  transaction triggers the entry criteria of your system first, then
>  within the same time-based bar a transaction triggers the exit criteria
>  of your system. TradeStation handles both signals in the order in which
>  the transactions are received on a real-time/delayed basis.
>  
>  If the exit signal of a trading system is triggered first in a
>  time-based bar in which both the entry and exit criteria of a system
>  are met, TradeStation handles the exit criteria in the following ways:
>  
>  If you do not have an open position and a transaction triggers the exit
>  criteria of your system first and then a later transaction triggers the
>  entry criteria, TradeStation ignores the exit criteria (because at the
>  time there is no open position to exit) but issues the entry signal to
>  open a position.
>  
>  If you do have an open position and on the same time-based bar a
>  transaction triggers the exit criteria of your system and then a
>  transaction triggers the entry criteria of your system, TradeStation
>  shows the open position as having been liquidated due to the exit
>  criteria, then would show a new open position based on the entry
>  signal.
>  
>  As you can see, when you apply a trading system to a time-based chart
>  based on real-time delayed data, TradeStation can calculate the entry
>  and exit criteria of your trading system with great precision and enter
>  and exit the market based on exactly what is happening in the market
>  when one or more of the criteria of your system are met.
>  
>  Understanding How TradeStation Handles Entry and Exit Orders on
>  Time-Based Bars Built with Historical Data
>  
>  The information in the previous heading provided a frame of reference
>  for you by covering the way TradeStation handles the entry and exit
>  signals generated by a trading system when applied to a time-based
>  chart based on real-time/delayed data.
>  
>  Before you can understand the way TradeStation simulates market
>  activity when you apply a trading system to a chart of time-based bars
>  based on historical data, you need to understand why it is necessary
>  for TradeStation to simulate market activity when you apply a trading
>  system to a chart for time-based bars based on historical data.
>  
>  When you work with time-based bars based on historical data,
>  TradeStation cannot know the chronological order of the transactions
>  that make up the bar. The only transactions we can be sure of with
>  time-based bars based on historical data are the open, which occurred
>  first, and the close, which occurred last. With time-based bars based
>  on historical data there is no way to be sure whether the market opened
>  and then went down, or the market opened and then went up.
>  
>  However, because of the importance of the order of the ticks when you
>  have applied a trading system to time-based bars based on historical
>  data, Omega Research spent a great deal of time studying the movement
>  of various markets to discover a way to determine the most likely
>  chronological order of the high and low ticks.
>  
>  After extensive research, a general rule was established about the
>  chronological order in which ticks occur. Under this rule, if the open
>  of the bar is closer to the low of the bar than to the high of the bar,
>  TradeStation handles the bar as if the low was reached first and then
>  the price went to the high before moving to the close. If the open
>  price of the bar is closer to the high of the bar, then TradeStation
>  handles the bar as if the high was reached first, and the price went to
>  the low before moving to the close.
>  
>  The order of the transactions that make up the bar becomes very
>  important when you apply a trading system to time-based bars based on
>  historical data. For example, if you had no market position and the
>  entry and exit criteria of the trading system were both triggered on
>  the same bar, which signal was triggered first is a very important
>  issue.
>  
>  If the exit criteria of the system were triggered first, TradeStation
>  would ignore the exit signal (because there was no market position to
>  exit) and, when the entry signal was triggered on the same bar,
>  TradeStation would issue the entry order, showing you as having opened
>  a new position.
>  
>  
>  
>  Order in which transactions are assumed to have been made
>  
>  The bars in the above figure show the order in which TradeStation would
>  assume transactions had occurred. In the above figure, transactions are
>  marked 1, 2, 3 and 4. The open price of a bar is always number 1. The
>  close price of the bar is always number 4. The numbers 3 and 4 are used
>  to designate the most likely chronological order of the high and low
>  ticks.
>  
>  Rarely does the market strictly follow the 1, 2, 3, 4, order
>  illustrated in the previous figure. Instead, the market bounces up and
>  down. For example, even on a day when the market climbs in a relatively
>  steady manner, there are brief dips followed by still higher highs.
>  
>  When you apply a trading system to a chart based on historical data, a
>  special TradeStation innovation called Bouncing Ticks was created to
>  simulate the irregular up and down way the market really moves, even
>  during an up-trend or a down-trend.
>  
>  The way the Bouncing Ticks function works is simple enough to
>  understand: When market conditions trigger the entry signal of your
>  trading system (either long or short), TradeStation bounces back by a
>  particular percentage that you can specify. If, within that percentage,
>  there is a price that would trigger the exit criteria of your system,
>  TradeStation exits the position on the same bar on which you entered
>  the position.
>  
>  Here is the major reason that TradeStation simulates market activity in
>  this way: During actual market activity, chances are extremely high
>  that once the entry signal of your trading system is triggered the
>  market would turn against you by a certain percentage. When testing a
>  trading system on historical data, you need to take this normal
>  bouncing market activity into account, particularly when you design
>  your system's exit criteria. For example, in real-time market
>  conditions, when the market turns against you, you have no way of
>  knowing if the market will rebound in your direction. Your system
>  should have been constructed in such a way that it exits your position
>  before you sustain losses that you cannot afford.
>  
>  When testing a trading system on a time-based chart based on historical
>  data, it can signal a flaw in your trading system if the system
>  repeatedly enters and exits on the same bar. In such a case, we
>  recommend that you re-examine the rules of your system and redesign the
>  entry and exit criteria so that the exit criteria of the system
>  liquidates your open position only when the market takes a significant
>  turn against you.
>  
>  Setting the Bouncing Ticks Option
>  
>  As detailed in the previous heading, TradeStation uses an exclusive
>  feature called Bouncing TicksTM to simulate actual market activity when
>  you apply a trading system to time-based bars based on historical data.
>  The percentage that TradeStation bounces back in order to simulate
>  market activity is, by default, 10 percent of the total price range of
>  the bar. For example, if the high of the bar were 100 and the low were
>  90, TradeStation would bounce back one point whenever the entry signal
>  of a trading system was triggered.
>  
>  The following provides directions for changing the default percentage
>  that TradeStation bounces back when the entry criteria of a trading
>  system are triggered when you apply a trading system to time-based bars
>  based on historical data.
>  
>  1. Use the Tools - Options menu sequence to open the Options dialog. 2.
>  Click the System tab to produce the System dialog. 3. In the Percent
>  increment for Bouncing Ticks edit box, delete the
>     current value (by default, 10 percent) and enter a new value.
>  4. Click OK to return to TradeStation.
>  
>  Omega Research decided on the default of 10 percent after extensive
>  research into how the markets move during the trading day. We recommend
>  that you not change the default unless you are testing the system to
>  determine how it performs in what might be considered non-typical
>  market conditions.
>  
>  See Also:
>  
>  Understanding Trading System Signals in a Chart
>  Using the System Report to View System Results
>  Using the System Equity Indicator During Testing
>  Specifying Trading System Input Values
>  Adjusting Trading System Profit-Loss Results for Costs
>  Using Stops to Limit Your Potential Losses
>  Setting Entry Limits on Trading Systems
>  Understanding Trading System Order Types
>  Specifying Trailing Stops
>  Deleting and Turning-Off Systems
>  
>  Formatting Trading System Colors and Style
>  
>  Other relevant chapters are:
>  
>  Chapter 13, Getting to Know Trading Systems
>  
>  Chapter 15, Optimizing Your Trading Systems
>  Chapter 16, Automating Your Trading Systems