The standard RSI is added. If the value of the indicator is less than BuyOp and the current value of the indicator is greater than the previous one then we buy. If the value of the indicator is greater than SellOp and the current value is less than the previous one then we sell. Test is the period of the RSI. Trailing Stop that is used here was taken either from this forum or from the Alpari forum (I don't remember exactly). The auto-optimizer is added from the article: Automated Optimization of a Trading Robot in Real Trading.
The parameters for the optimization are: BuyOp, SellOp,Test.
The chart is only for one day because the parameters are optimized everyday.
It behaves very good on the M1. The best values are on EURJPY.
|Symbol||EURJPY (Euro vs Japanese Yen)|
|Period||1 Minute (M1) 2008.10.17 00:00 - 2008.10.17 22:59 (2008.10.17 - 2008.10.20)|
|Model||Every tick (the most precise method based on all available least timeframes)|
|Parameters||TakeProfit=50; Lots=0.1; RiskPercentage=10; TrailingStop=50; MaxOrders=1; BuyOp=29; SellOp=74; magicnumber=777; Test=11; SetHour=0; SetMinute=10;|
|Bars in test
|Missmatched charts errors||0|
|Total net profit
||87.46 (17.25%)||Relative drawdown
||5||Short positions (% won)||3 (100.00%)||Long positions (% won)||2 (100.00%)|
|Profit trades (% of total)||5 (100.00%)||Loss trades (% of total)||0 (0.00%)|
|Largest||profit trade||52.07||loss trade||0.00|
|Average||profit trade||50.89||loss trade||0.00|
|Maximum||consecutive wins (profit in money)||5 (254.46)||consecutive losses (loss in money)||0 (0.00)|
|Maximum||consecutive profit (count of wins)||254.46 (5)||consecutive loss (count of losses)||0.00 (0)|
|Average||consecutive wins||5||consecutive losses||0|
|№||Time||Type||Order||Volume||Price||S / L||T / P||Profit||Balance|
It doesn't take part in the Championship because of the auto-optimization (it can be used under the Championship, and the code is not mine as well), and the idea came to me after the beginning of the Championship.
Translated from Russian by MetaQuotes Software Corp.
Original code: https://www.mql5.com/ru/code/8513
Automatic Trading SystemStochastic Net
Stochastic net for the the classification problems with the instruction provided.
It displays the current time position in relation to the beginning and the end of the bar. In addition the ratio between the time passed since the beginning of the bar and the whole bar duration expressed in percents is given. The informer is useful for cSwetten
EA with a neuro net (a net with error back propagation). These are the firsts steps. I upload it for the intelligent criticism.