Back Testing Accuracy with OHLC data

Discussion in 'Strategy Building' started by SimpleMeLike, Feb 23, 2017.

  1. Initially, it was unclear, how your signals were being generated. From your example the signal occurs at the CLOSE. This is SAFE, as it is a known point in time, so using any data from that BAR was fine (as it would be present or past). -- The concern (references to Cheating) would only come into play, if the signal were at an indeterminate time point, say near a High or a Low. My reference to "SIGNAL" is meant to refer to the point "if bar open and closes above line1 (line1 is some calculated price)" in your reference above! This is the point a decision is made to place an order! (A fill would be a subsequent event which follows the signal)! --

    Since the note: "Strategies will be processed at the close of each bar" <-- You should not have issue with the referenced form of "cheating", as the time of the signal is at CLOSE!
     
    #21     Feb 24, 2017
  2. Thanks stepandfetchit for responding.

    According to Ninja Trader, "During backtest, strategies can ONLY be processed at the close of each bar" See the attachment.

    Screenshot_1.png

    I also called Ninja Trader support regarding the same example i ask you during back test and they say the accuracy issue of base on bar type as well. But my example of limit fill should be filled, but there back testing does not display real live trading cause of back testing does not perform the dynamic of market conditions during the back testing.

    I believe I understand better now the back testing now.

    For me, I want to make sure I understand any flaw of what i am doing. I don't like making technical mistakes any more because of lack of knowledge.
     
    #22     Feb 24, 2017