Multi Layer Perceptron Stat Arb: Free MT5 Expert Advisor

Overview

Multi Layer Perceptron Stat Arb periodically scans major currency pairs to identify high correlation relationships across multiple lookback windows. Trade signals are generated based on a confidence threshold derived from neural network scoring. The EA features risk controls including fixed lot sizing, USD-based stop loss and take profit levels, and a maximum bar-based hold time. Note that the included neural network weights are placeholders requiring external training or optimization.

Multi-layer perceptron neural network scoring

Automated currency pair correlation scanner

Time-lagged correlation analysis

USD-denominated risk management

Time-based trade exit logic

Details

Author

4SIGHTalgo on GitHub

Type

Trading robot

Platform

MetaTrader 5 (MQL5)

Strategy

Arbitrage, AI / machine learning, Mean reversion

Markets

Forex

Timeframes

Any timeframe

Risk profile

No grid, martingale or hedging logic was found in its code or README. That says nothing about how it performs: backtest it first.

EA files

Source

github.com/4SIGHTalgo/MQL5-Mutli-Layer-Perception

Input parameters

Read from MLP_Stat_Arb.mq5. Defaults are the author's; change them in the EA's Inputs tab.

Name	Type	Default	Description
`InpMinCorrelationForTrading`	`double`	`0.7`	Min abs correlation to consider a pair tradeable
`InpScannerUpdateHours`	`int`	`4`	How often to re-scan all pairs (in hours)
`InpLookbackShort`	`int`	`20`	Short-term analysis window
`InpLookbackMid`	`int`	`50`	Medium-term analysis window
`InpLookbackLong`	`int`	`100`	Long-term analysis window
`InpLagPeriod`	`int`	`4`	Lag for the time-lagged correlation
`InpScalingLookback`	`int`	`200`	Bars to look back for Min/Max feature scaling
`InpLotSize`	`double`	`0.1`	Lot size for each trade
`InpMagicNumber`	`int`	`3007`	Magic number for trades
`InpTakeProfitUSD`	`double`	`1000`	Take Profit for the spread in account currency (e.g., USD)
`InpStopLossUSD`	`double`	`1000`	Stop Loss for the spread in account currency (e.g., USD)
`InpMaxHoldBars`	`int`	`24000`	Max number of bars to hold a trade

Show all 36 inputs

Name	Type	Default	Description
`InpCorrelationExitThreshold`	`double`	`0.3`	Exit if long-term Pearson correlation drops below this
`InpMLPSignalThreshold`	`double`	`1.2`	Confidence level to open a trade
`inp_W1_11`	`double`	`0.0, inp_W1_12=0.0, inp_W1_13=0.0, inp_W1_14=0.0, inp_W1_15=`
`inp_W1_21`	`double`	`0.0, inp_W1_22=0.0, inp_W1_23=0.0, inp_W1_24=0.0, inp_W1_25=`
`inp_W1_31`	`double`	`0.0, inp_W1_32=0.0, inp_W1_33=0.0, inp_W1_34=0.0, inp_W1_35=`
`inp_W1_41`	`double`	`0.0, inp_W1_42=0.0, inp_W1_43=0.0, inp_W1_44=0.0, inp_W1_45=`
`inp_W1_51`	`double`	`0.0, inp_W1_52=0.0, inp_W1_53=0.0, inp_W1_54=0.0, inp_W1_55=`
`inp_W1_61`	`double`	`0.0, inp_W1_62=0.0, inp_W1_63=0.0, inp_W1_64=0.0, inp_W1_65=`
`inp_b1_1`	`double`	`-0.4, inp_b1_2= 0.7, inp_b1_3=-0.9, inp_b1_4= 0.6, inp_b1_5=`
`inp_W2_11`	`double`	`0.0, inp_W2_12=0.0, inp_W2_13=0.0, inp_W2_14=0.0, inp_W2_15=`
`inp_W2_21`	`double`	`0.0, inp_W2_22=0.0, inp_W2_23=0.0, inp_W2_24=0.0, inp_W2_25=`
`inp_W2_31`	`double`	`0.0, inp_W2_32=0.0, inp_W2_33=0.0, inp_W2_34=0.0, inp_W2_35=`
`inp_W2_41`	`double`	`0.0, inp_W2_42=0.0, inp_W2_43=0.0, inp_W2_44=0.0, inp_W2_45=`
`inp_W2_51`	`double`	`0.0, inp_W2_52=0.0, inp_W2_53=0.0, inp_W2_54=0.0, inp_W2_55=`
`inp_W2_61`	`double`	`0.0, inp_W2_62=0.0, inp_W2_63=0.0, inp_W2_64=0.0, inp_W2_65=`
`inp_W2_71`	`double`	`0.0, inp_W2_72=0.0, inp_W2_73=0.0, inp_W2_74=0.0, inp_W2_75=`
`inp_W2_81`	`double`	`0.0, inp_W2_82=0.0, inp_W2_83=0.0, inp_W2_84=0.0, inp_W2_85=`
`inp_b2_1`	`double`	`-1.0, inp_b2_2= 0.2, inp_b2_3= 0.9, inp_b2_4=-0.7, inp_b2_5=`
`inp_W3_11`	`double`	`0.0, inp_W3_12=0.0, inp_W3_13=0.0`
`inp_W3_21`	`double`	`0.0, inp_W3_22=0.0, inp_W3_23=0.0`
`inp_W3_31`	`double`	`0.0, inp_W3_32=0.0, inp_W3_33=0.0`
`inp_W3_41`	`double`	`0.0, inp_W3_42=0.0, inp_W3_43=0.0`
`inp_W3_51`	`double`	`0.0, inp_W3_52=0.0, inp_W3_53=0.0`
`inp_b3_1`	`double`	`0.8, inp_b3_2=-0.6, inp_b3_3= 0.8`

How to install it

1Download the sourceUse "Download source (.zip)" above, or clone 4SIGHTalgo/MQL5-Mutli-Layer-Perception from GitHub, and unzip it.

2Open the data folderIn MetaTrader choose File, then Open Data Folder, and go to MQL5/Experts.

3Copy the filesCopy the .mq5 file into Experts.

4Compile in MetaEditorOpen the file in MetaEditor and press Compile. Fix any missing includes the compiler reports.

5Test before you tradeRun it in the Strategy Tester, then attach it to a demo chart and allow algorithmic trading.

Third-party code. ForexR did not write, test or endorse this EA, and the summary above was generated automatically from its README and code. ForexR lists only repositories with readable source code and never hosts files. Read the code, backtest it and use a demo account first; automated trading can lose money quickly.

Similar expert advisors

AI Forex Istoch BB CounterTrend

by marco82ger

This MetaTrader 5 expert advisor executes a counter-trend strategy utilizing Bollinger Bands and the Slow Stochastic oscillator. Developed as part of academic research into deep learning and technical analysis, the robot incorporates Heiken Ashi filtering to identify market extensions and doji candles. It manages risk

MT5 Trading robot Mean reversionAI / machine learning

Updated 4 years ago None stated GitHub

Heiken Ashi Complete Strats

by jsgaston

This MetaTrader 5 expert advisor provides a versatile framework for trading based on Heiken Ashi price action. It allows users to toggle between numerous built-in strategies, including trend following, moving average crossovers, Fibonacci levels, and various oscillators like RSI and MACD. The system includes an option

MT5 Trading robot Trend followingIndicator crossoverMean reversion

Updated 2 years ago None stated GitHub

Gold Futures Spot Arbitrage

by HeYuansong

This expert advisor facilitates arbitrage trading between futures and spot instruments, primarily targeting gold and stock indices. Available for MetaTrader 4 and 5, it monitors the price difference between two related symbols against a moving average to identify trade opportunities. The logic includes a noise-filterin

MT4MT5 Trading robot ArbitrageMean reversionGrid High risk

Updated 3 years ago None stated GitHub

Linear Regression Trading Bot

by nickmaglowsch

This MetaTrader 5 expert advisor implements a trade execution strategy based on a linear regression model. It calculates entries using specific coefficients for an intercept and five independent variables, which include volatility, relative strength index, and average directional index data. The bot supports intraday o

MT5 Trading robot AI / machine learningMean reversion

Updated 5 years ago None stated GitHub

AI Quantum Learning Advanced Trailing Stop

by GBNarasimha

The AI Quantum Learning Advanced Trailing Stop EA is an automated trading robot for MetaTrader 5 that employs neural network structures and quantum probability calculations. It monitors market patterns by analyzing price changes, volume, volatility, and indicators like RSI. The EA manages risk using fixed stop loss and