Nirvana

Hi Community.

I´m requesting a rather complex EA. Better saying, it´s a mix of EA´s.

The EA should choose between Strategies, according to market conditions (https://www.mql5.com/en/articles/350) ; use Suport Vector Machine to screen the signal from the Indicators of the Strategies (https://www.mql5.com/en/articles/584); Optimize Indicators if a certain Drawdown is reached  or if a certain count of "False " signals is given by Suport Vector Machine (https://www.mql5.com/en/articles/350).

Also there should be a feature "Add Positions", according to user defined parameters.

Attached, there´s a Fluxogram or Algorithm of how the EA should behave.

Any further information, just feel free to ask.

Full description

Open Trade Strategies

There should be 4 different Trading Strategies, according to 4 sets of Indicators.

Strategy 1

-Stochastic, Fractal, BW´s Accelerator/Deaccelerator Oscilator, BW´s Awsome Oscilator

Strategy 2

-Moving Averages, Fractal, BW´s Accelerator/Deaccelerator Oscilator, BW´s Awsome Oscilator

Strategy 3

-Parabolic Sar, Fractal, BW´s Accelerator/Deaccelerator Oscilator, BW´s Awsome Oscilator

Strategy 4

-Envelopes, Fractal, BW´s Accelerator/Deaccelerator Oscilator, BW´s Awsome Oscilator

First Steps 

The EA must support 4/5 digits Brokers, as well as ECN Brokers. 

The EA should Scan/Monitor the Instrument on the current Time Frame to which chart it is attached, not all Instruments on Market Watch.

When the EA is attached to a chart for the first time, it should optimize the Indicators from the different 4 Strategies using Genetic Algorithm, as referenced on article (https://www.mql5.com/en/articles/350).  

The Indicators to optimize should be only: Stochastic, from Strategy 1; Moving Averages, from Strategy 2; Parabolic Sar, from Strategy 3; and Envelopes, from Strategy 4. No need to optimize all other Indicators. 

The EA then should then use Support Vector Machine to, according to a certain elapsed time defined by user, read and "learn" the tick data from the Instrument to which chart is attached. Support Vector Machine learning tool and its usage, is described on article (https://www.mql5.com/en/articles/584) . 

After these steps are done, the EA should scan/monitor the instrument to which it is attached, using the 4 different strategies, waiting for a signal from one of them. 

The first strategy to trigger a signal, would be the one to be used to open a trade. 

As soon there´s an open trade, no need for more monitoring the instrument of the chart by the Strategies, avoiding this way too much resources consuming (Memory) of the PC by the MT5  

Every time a strategy triggers a signal (buy/sell), it should be screened/tested by the Support Vector Machine learning tool. And every signal should be screened/tested by a different Support Vector Machine: one for buy and one for sell.   

Money Management

You should consider these conditions for Money Management

-0.01 for each 1000$ on equity

-Lot volume should be calculated in "x" percentage of the Free Margin ("x" user defined)

-Don´t add new trades/new positions if Margin Level(%)<"x" ("x" user defined)

-Initial Stop Loss when opening a new trade, should be "x" fractals behind, and on a selectable time frame. And "x" must be an input parameter, defined by user.

-Money Magement should have a Candle Size Filter input, meaning that a signal triggered (by the Strategies or by the Add Positions feature) right after a candle with a size of "x" Pips, the signal from it should be ignored. And the "x" value in Pips for the Candle Size should be user defined. 

Time Schedule

-The user should be able to define on which between hours the EA can trade, as well as the days of the week.

Adding Positions Feature

This feature adds a new position to an existing trade,  using a signal from these 3 Indicators: Fractal, BW´s Accelerator/Deaccelerator Oscilator and BW´s Awsome Oscilator .

Every time these indicators trigger a combined signal on the same direction of the trade, it should add a new position to an existing trade, according to Money Management. 

Regard that:

-This feature should be turned on/off by user

-This feature never opens a new trade

-The user should be able to choose from 1 to 3 of these Indicators, and each one should be Time Frame selectable 

-Every signal is filtred/screened by a different Support Vector Machine learning tool: one for Buy and one for Sell. 

Closing a Trade  (Trailing Stop/Stop Loss/Take Profit)

The EA should be able to use one or both of these ways of trailing/take profit : Parabolic SAR and Opposite Signal  from a set of 3 Indicators

-You should consider as Opposite Signal, a combined signal on the opposite direction of the trade, from Indicators : Fractal, BW´s Accelerator/Deaccelerator Oscilator and BW´s Awsome Oscilator 

-The user should be able to choose from one 1 to 3 of these Indicators, and each one should be Time Frame selectable 

Optimizing Indicators

The Indicators of the Strategies (Stochastic, Moving Averages, Parabolic Sar and Envelopes) should allways be optimized by the Genetic Algorithm every time:

-The EA is attached for the first time to a chart

-Every time a trade has reached a certain "x" percentage of drawdown (the value "x" should be user defined)

-The Support Vector Machine has returned a "False" value for a certain "x" times ("x" value should be user defined), from the signals Inputs (buy/sell) that were given by the Strategies

-After a number "x" of trade/deals have been done by the EA ("x" value should be user defined)

Logging

-There must be a Log for the output values (True/False) of each Support Vector Machine, for better solving any issue regarding the input signal (Buy/sell) given by the different Strategies.

A lot of the code is available as open source code on the webpages of the described articles above, so please be fair and honest with your budgets. 

Any doubts or questions, just feel free to ask. 

- 

-