Fractal breakout intraday Strategy EUR/USD 1H –

Forums › ProRealTime English forum › ProOrder support › Fractal breakout intraday Strategy EUR/USD 1H –

Viewing 15 posts - 241 through 255 (of 360 total)
1 2 3 4 5 … 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
  • 09/26/2017 at 3:24 PM #47413
    CKW

    OK. re-attached

    09/27/2017 at 8:41 AM #47462
    Nicolas

    “0 bar” mean that the trade has been opened and closed in the same bar, with tick mode enable, each tick of the bar is tested, so the backtest should be accurate.

    09/27/2017 at 3:56 PM #47540
    stefou102

    Hello,

    I have taken the version from CKW and tried to improve, so that the code works also with other currencies (with different optimized parameters). I added:

    • a closure of all opened positions before WE (avoiding Sunday gap)
    • an additional filter based on Stochastics.
    • additional filter for the short side (same as long side already introduced by CKW)
    • I replaced the trailing stop by the one developed by Nicolas

    Compared to previous version on EURUSD, backtest takes even less positions (so lower total return), but shows superior Gain/Loss ratio (>4!!). Other currencies works also, even if the KPI are a bit less impressive, although good enough to trade imo. Maybe the code are too optimized, I only have 3 years of 1H historical data. For those who have more, they can maybe run some more robust backtests?

    What is your opinion, readers?

    1 user thanked author for this post.
    avatar victormork
    10/02/2017 at 8:56 AM #47967
    victormork

    Hi Stefou,

    Nice work with trying some additional pairs. I my opinion is EURGBP the best addition.

    If I could make one suggestion it would be to try changing all targets and “pip based” commands to be ATR based which would avoid fitting values to current market conditions.

    10/02/2017 at 9:19 AM #47970
    ALE

    Hello

    as we could see in the last months , many results are different from backtest.

    we should find dome different way to avoid this problem

     

    some idea?

     

    10/02/2017 at 9:28 AM #47974
    victormork

    Hi, I think it would be good to try a greater distance for trailing stops and profit targets. You might get a lower win rate but that’s irrelevant because you will be able to ride winners in a much more efficient way. Maybe also try using ATR to determine the distance?

    10/02/2017 at 10:16 AM #47978
    Despair

    @ALE: Which kind of difference between live and backtest did you discover?

    10/02/2017 at 11:28 AM #47984
    ALE

    X example this is an immane of last month backtest vs real

    10/02/2017 at 12:25 PM #47995
    Despair

    Wow, this IS a difference! Did you have a look at single trades? What happens there exactly? It looks like for the live version the trades are stopped out pretty early. Are you sure both versions have the same stop loss?

    10/02/2017 at 3:15 PM #48015
    CKW

    My edited version don’t have major significant difference. Except trade on 5-Sep triggered in live but not in backtest.

    10/02/2017 at 3:32 PM #48020
    CKW

    Hello Stefou,

    I like Your “improved version”, both GL for short & buy more than 2.5. That’s great.
    I started yours in demo live:)
    thanks for your sharing!

    10/02/2017 at 8:37 PM #48061
    stefou102

    Hello CKW you may want to edit the code where we exit all positions before WE .  I know use this, to adjust when daylight changes (a copy from the forum):

    // quit all positions before we
    if (currentmonth = 3 and day >= 15) or (currentmonth = 11 and day < 8) then //daylight-saving corrections
    dlc = 1
    else
    dlc =0
    endif
    //no pos before WE
    If dayofweek = 5 and hour >= 22 -dlc Then
    If longonmarket Then
    Sell at market
    ElsIf shortonmarket Then
    Exitshort at market
    EndIf
    EndIf

    10/04/2017 at 5:43 PM #48213
    Vonasi

    I was trying out a strategy idea last week using the universal/standard Bill Williams fractal code found here in the library but I stopped development on the strategy when I noticed that the fractal levels calculated in the strategy code were not the same as those shown by an indicator with identical fractal code applied on the chart. I’ve just applied the same indicator with ALES fractal breakout strategy and noticed the levels also do not match. Anyone else noticed this?

    10/04/2017 at 6:15 PM #48215
    ALE

    @ Vonasi

    I’ve change level from High/low to close, because I want to consider close level breakout

     

    //BILL WILLIAM FRACTAL INDICATOR
    //CP=PERIOD
    CP=113
    if close[cp] >= highest[2*cp+1](close) then
    LH= 1
    else
    LH=0
    endif
    if close[cp] <= lowest[2*cp+1](close)  then
    LL= –1
    else
    LL=0
    endif
    if LH=1 then
    HIL= close[cp]
    endif
    if LL  = –1 then
    LOL=close[cp]
    endif
    10/04/2017 at 7:07 PM #48217
    Vonasi

    Aaahh – that explains the difference. Any particular logic behind using CLOSE rather than LOW/HIGH? I believe that you ‘d get more false breakouts but then again also get earlier entry. Have you tried it with TYPICALPRICE rather than CLOSE?

  • Author
    Posts
Viewing 15 posts - 241 through 255 (of 360 total)
1 2 3 4 5 … 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Find exclusive trading pro-tools on ProRealCode Marketplace

Similar topics:
bollinger+fractal strategy idea
breakout box strategy (tick by tick not working for entries)
entre 20h30 et 21h00, acheter 1 contrat au marché
eur/usd backtest spread

Create your free account now and post your request to benefit from the help of the community
Register or Login