How to use tick data in a 2min system backtest? (Tradestation). please help.

Discussion in 'Strategy Development' started by piyayo, May 8, 2012.

  1. piyayo

    piyayo

    Well, I would appreciate any help about this, please!.

    I have a few automated systems on Tradestation designed for 2 min charts.

    I think that my backtest isn´t correct because: how does the program knows what is exactly the price when the system gives a signal to enter at market?.

    I mean, in real time you see the information comming tick by tick, but in a 2 min chart there is no such precission, right?

    How can you solve this on Tradestation?

    any suggestion about any other platform that could use tick by tick information on backtesting X min charts?

    Please help!

    Many thanks in advance.
     
  2. Tick-by-tick simulation

    We call this feature the Bar Magnifier. It is essential for increasing precision during backtesting. MultiCharts can construct larger bars out of smaller components—second and minute bars out of ticks, hour and day bars out of minutes. You can recreate exact price movements within each bar by using the Bar Magnifier, which will build larger bars out of smaller components. For example, one-hour bars have four visual points—open, high, low, and close. The Bar Magnifier can invisibly load minutes that make up the hour, and strategy will be backtested on a minute-by-minute basis.

    www.multicharts.com
     
  3. piyayo

    piyayo

    Many thanks,

    As I´m using tradestation by now, anyone knows how to do it in tradestation?.

    It has a "Use look-inside-bar-back-testing" feature, but anyone kows if this offers real and reliable results?

    I will try with multicharts also. Thanks for the information
     
  4. Tradestation 9.0 has look inside bar back testing with tick resolution.
     
  5. How important all this is (in terms of helping you better understand your strategy's performance) can be assessed by comparing the average holding period of each trade with the bar timeframe.

    For example, if your system's average holding period is say 4 hours, IMO you'll get little practical benefit out of switching from 2-min bars to ticks.

    If, on the otherhand, your average trade lasts two minutes, then even tick-based (i.e. "tick" = last traded price in stocks / last quote change in FX) backtesting probably isn't going to help you, as order queue position is going to be important and ticks don't tell you anything about that. You can make assumptions like "only execute a trade if price moves at least one tick beyond my trade price", but that will (in my experience) give you overly pessimistic results that don't really tell you how the strategy will do in real life. To get further in this case, you'll need full order book info (all executes and quote add/changes/cancels at all levels), and a simulator able to use the data.