English
View on GitHub

Five MA Multi-Timeframe Strategy

Overview

The Five MA Multi-Timeframe Strategy replicates the original MT4 "5matf" expert advisor using StockSharp's high-level API. The strategy analyzes five simple moving averages across three timeframes (primary, higher, and slowest) and combines the slope of each average with the Accelerator Oscillator to produce graded entry signals. When enough bullish or bearish evidence is present on all timeframes, the strategy opens or closes positions accordingly.

Indicators and Data

  • Simple Moving Averages (SMA): Periods 5, 8, 13, 21, and 34 on all three timeframes.
  • Accelerator Oscillator (AC): Applied on the primary and tertiary timeframes to assess momentum acceleration.
  • Timeframes: Default set to 15 minutes (signal), 60 minutes (confirmation), and 240 minutes (trend filter). All timeframes can be adjusted via parameters.

Signal Logic

  1. Each SMA compares its current value to the previous candle to determine an upward or downward slope.
  2. Accelerator Oscillator checks for bullish or bearish sequences using the latest four values.
  3. Slope counts and oscillator contributions are aggregated into percentage scores for every timeframe.
  4. When all three timeframes have bullish scores above 50%, a BUY signal is generated. Scores above 75% strengthen the signal.
  5. The same thresholds applied in the opposite direction generate SELL signals.
  6. Positions are closed when an opposite signal exceeds the configured close level. New trades only open when no position is active, mirroring the original expert advisor behavior.

Parameters

Name Default Description
CandleType 15-minute candles Primary timeframe used for trading signals.
HigherTimeframe1 60-minute candles First higher timeframe for confirmation.
HigherTimeframe2 240-minute candles Second higher timeframe for slow trend filter.
FirstPeriod – FifthPeriod 5, 8, 13, 21, 34 SMA lengths applied to each timeframe.
OpenLevel 0 Minimum signal grade required to open a new position.
CloseLevel 1 Opposite signal grade required to close an existing position.

All parameters can be optimized or fine-tuned within StockSharp's strategy UI.

Usage Notes

  • The strategy uses market orders and does not issue simultaneous reversals; it always waits for a flat position before opening in the opposite direction.
  • Enable history data feeds for all selected timeframes to ensure synchronized calculations.
  • Consider tuning the SMA lengths or oscillator usage when applying the strategy to different markets or volatility regimes.

Conversion Notes

This implementation retains the core behavior of the MT4 "5matf" expert advisor while leveraging StockSharp's subscription and indicator binding system. The accelerator logic requires four completed candles before signals become active, just like the original script.

using System;
using System.Collections.Generic;
using System.Linq;

using Ecng.Common;

using StockSharp.Algo.Strategies;
using StockSharp.BusinessEntities;
using StockSharp.Messages;

namespace StockSharp.Samples.Strategies;

/// <summary>
/// Five moving averages across three timeframes with Accelerator Oscillator confirmation.
/// </summary>
public class FiveMaMultiTimeframeStrategy : Strategy
{
	private readonly StrategyParam<DataType> _candleType;
	private readonly StrategyParam<DataType> _higherTimeframe1;
	private readonly StrategyParam<DataType> _higherTimeframe2;
	private readonly StrategyParam<int> _firstPeriod;
	private readonly StrategyParam<int> _secondPeriod;
	private readonly StrategyParam<int> _thirdPeriod;
	private readonly StrategyParam<int> _fourthPeriod;
	private readonly StrategyParam<int> _fifthPeriod;
	private readonly StrategyParam<int> _openLevel;
	private readonly StrategyParam<int> _closeLevel;

	private readonly Dictionary<string, FrameState> _frames = new(StringComparer.Ordinal);

	public DataType CandleType { get => _candleType.Value; set => _candleType.Value = value; }
	public DataType HigherTimeframe1 { get => _higherTimeframe1.Value; set => _higherTimeframe1.Value = value; }
	public DataType HigherTimeframe2 { get => _higherTimeframe2.Value; set => _higherTimeframe2.Value = value; }
	public int FirstPeriod { get => _firstPeriod.Value; set => _firstPeriod.Value = value; }
	public int SecondPeriod { get => _secondPeriod.Value; set => _secondPeriod.Value = value; }
	public int ThirdPeriod { get => _thirdPeriod.Value; set => _thirdPeriod.Value = value; }
	public int FourthPeriod { get => _fourthPeriod.Value; set => _fourthPeriod.Value = value; }
	public int FifthPeriod { get => _fifthPeriod.Value; set => _fifthPeriod.Value = value; }
	public int OpenLevel { get => _openLevel.Value; set => _openLevel.Value = value; }
	public int CloseLevel { get => _closeLevel.Value; set => _closeLevel.Value = value; }

	public FiveMaMultiTimeframeStrategy()
	{
		_candleType = Param(nameof(CandleType), TimeSpan.FromMinutes(15).TimeFrame()).SetDisplay("Candle Type", "Primary signal timeframe.", "Timeframes");
		_higherTimeframe1 = Param(nameof(HigherTimeframe1), TimeSpan.FromMinutes(60).TimeFrame()).SetDisplay("Higher Timeframe 1", "First confirmation timeframe.", "Timeframes");
		_higherTimeframe2 = Param(nameof(HigherTimeframe2), TimeSpan.FromMinutes(240).TimeFrame()).SetDisplay("Higher Timeframe 2", "Slow trend timeframe.", "Timeframes");

		_firstPeriod = Param(nameof(FirstPeriod), 5).SetGreaterThanZero();
		_secondPeriod = Param(nameof(SecondPeriod), 8).SetGreaterThanZero();
		_thirdPeriod = Param(nameof(ThirdPeriod), 13).SetGreaterThanZero();
		_fourthPeriod = Param(nameof(FourthPeriod), 21).SetGreaterThanZero();
		_fifthPeriod = Param(nameof(FifthPeriod), 34).SetGreaterThanZero();

		_openLevel = Param(nameof(OpenLevel), 0).SetNotNegative().SetDisplay("Open Level", "Minimum signal grade required to open.", "Signal");
		_closeLevel = Param(nameof(CloseLevel), 1).SetNotNegative().SetDisplay("Close Level", "Opposite signal grade required to close.", "Signal");
	}

	public override IEnumerable<(Security sec, DataType dt)> GetWorkingSecurities()
		=> [(Security, CandleType), (Security, HigherTimeframe1), (Security, HigherTimeframe2)];

	protected override void OnReseted()
	{
		base.OnReseted();
		_frames.Clear();
	}

	protected override void OnStarted2(DateTime time)
	{
		base.OnStarted2(time);
		StartFrame("primary", CandleType, true);
		StartFrame("higher1", HigherTimeframe1, false);
		StartFrame("higher2", HigherTimeframe2, true);
	}

	private void StartFrame(string key, DataType candleType, bool useAc)
	{
		var frame = new FrameState([FirstPeriod, SecondPeriod, ThirdPeriod, FourthPeriod, FifthPeriod], useAc);
		_frames[key] = frame;

		var subscription = SubscribeCandles(candleType);
		subscription.Bind(candle =>
		{
			if (candle.State != CandleStates.Finished)
				return;

			frame.Process(candle);
			Evaluate();
		}).Start();

		var area = CreateChartArea();
		if (area != null)
			DrawCandles(area, subscription);
	}

	private void Evaluate()
	{
		if (_frames.Count != 3 || _frames.Values.Any(f => !f.IsReady))
			return;

		var bullishGrade = _frames.Values.Min(f => Grade(f.BullishScore));
		var bearishGrade = _frames.Values.Min(f => Grade(f.BearishScore));

		if (Position > 0)
		{
			if (bearishGrade > CloseLevel)
				SellMarket(Math.Abs(Position));
			return;
		}

		if (Position < 0)
		{
			if (bullishGrade > CloseLevel)
				BuyMarket(Math.Abs(Position));
			return;
		}

		if (bullishGrade > OpenLevel && bullishGrade > bearishGrade)
			BuyMarket();
		else if (bearishGrade > OpenLevel && bearishGrade > bullishGrade)
			SellMarket();
	}

	private static int Grade(decimal score)
		=> score > 75m ? 2 : score > 50m ? 1 : 0;

	private sealed class FrameState(int[] periods, bool useAc)
	{
		private readonly int[] _periods = periods;
		private readonly bool _useAc = useAc;
		private readonly List<decimal> _closes = [];
		private readonly List<decimal> _medians = [];
		private readonly List<decimal> _ao = [];
		private readonly List<decimal> _ac = [];
		private decimal[] _previousMas;

		public decimal BullishScore { get; private set; }
		public decimal BearishScore { get; private set; }
		public bool IsReady { get; private set; }

		public void Process(ICandleMessage candle)
		{
			_closes.Add(candle.ClosePrice);

			if (_useAc)
			{
				_medians.Add((candle.HighPrice + candle.LowPrice) / 2m);

				if (_medians.Count >= 34)
				{
					var ao = AverageTail(_medians, 5) - AverageTail(_medians, 34);
					_ao.Add(ao);

					if (_ao.Count >= 5)
						_ac.Add(ao - AverageTail(_ao, 5));
				}
			}

			if (_closes.Count < _periods.Max())
				return;

			var mas = _periods.Select(p => AverageTail(_closes, p)).ToArray();

			if (_previousMas is null)
			{
				_previousMas = mas;
				return;
			}

			var bullishVotes = 0;
			var bearishVotes = 0;

			for (var i = 0; i < mas.Length; i++)
			{
				if (mas[i] > _previousMas[i])
					bullishVotes++;
				else if (mas[i] < _previousMas[i])
					bearishVotes++;
			}

			if (_useAc && _ac.Count >= 4)
			{
				var last = _ac.Count - 1;
				var bullishAc = _ac[last - 3] < _ac[last - 2] && _ac[last - 2] < _ac[last - 1] && _ac[last - 1] < _ac[last];
				var bearishAc = _ac[last - 3] > _ac[last - 2] && _ac[last - 2] > _ac[last - 1] && _ac[last - 1] > _ac[last];

				if (bullishAc)
					bullishVotes++;
				else if (bearishAc)
					bearishVotes++;
			}

			var total = _useAc ? 6m : 5m;
			BullishScore = bullishVotes / total * 100m;
			BearishScore = bearishVotes / total * 100m;
			IsReady = true;
			_previousMas = mas;

			var keep = Math.Max(_periods.Max(), 40);
			if (_closes.Count > keep)
				_closes.RemoveRange(0, _closes.Count - keep);
			if (_medians.Count > 40)
				_medians.RemoveRange(0, _medians.Count - 40);
			if (_ao.Count > 10)
				_ao.RemoveRange(0, _ao.Count - 10);
			if (_ac.Count > 10)
				_ac.RemoveRange(0, _ac.Count - 10);
		}

		private static decimal AverageTail(List<decimal> values, int count)
		{
			var start = values.Count - count;
			var sum = 0m;
			for (var i = start; i < values.Count; i++)
				sum += values[i];
			return sum / count;
		}
	}
}