日本語
GitHub で見る

線形クロス取引戦略

この戦略は出来高に基づいた価格の線形回帰を計算して予測価格を生成します。予測価格が加重移動平均線を上抜け、かつ MACD ラインがシグナルを上回って上昇しているときにロングポジションを建てます。MACD ラインがシグナルを下回って低下し、直近の安値が切り下がっているときにショートポジションを建てます。

詳細

  • エントリー条件:
    • ロング: 予測価格が WMA を上抜け、かつ MACD がシグナルを上回って上昇。
    • ショート: MACD がシグナルを下回って低下し、安値が切り下がる。
  • ロング/ショート: 両方向。
  • エグジット条件: なし。逆方向のシグナルでポジションを反転。
  • ストップ: いいえ。
  • デフォルト値:
    • Length = 21.
    • LinearLength = 9.
    • CandleType = TimeSpan.FromMinutes(5).TimeFrame().
  • フィルター:
    • カテゴリ: トレンド
    • 方向: 両方
    • インジケーター: Linear Regression, WMA, MACD
    • ストップ: いいえ
    • 複雑さ: 中程度
    • 時間軸: イントラデイ
    • 季節性: いいえ
    • ニューラルネットワーク: いいえ
    • ダイバージェンス: いいえ
    • リスクレベル: 中
using System;
using System.Collections.Generic;
using System.Linq;

using Ecng.Common;

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

namespace StockSharp.Samples.Strategies;

/// <summary>
/// Linear Cross Trading strategy.
/// Regresses the close on volume over Length candles and predicts the price for the current volume. A long opens when the
/// predicted price crosses above its LinearLength WMA while MACD is above its signal and rising. A short opens when MACD is
/// below its signal and falling while the low is lower than the previous low. Opposite signals reverse the position.
/// </summary>
public class LinearCrossTradingStrategy : Strategy
{
	private readonly StrategyParam<int> _length;
	private readonly StrategyParam<int> _linearLength;
	private readonly StrategyParam<DataType> _candleType;

	private readonly Queue<(decimal close, decimal volume)> _window = new();
	private WeightedMovingAverage _wma;
	private decimal? _prevPredicted;
	private decimal? _prevWma;
	private decimal? _prevMacd;
	private decimal? _prevLow;

	/// <summary>
	/// Number of candles in the price-on-volume regression.
	/// </summary>
	public int Length
	{
		get => _length.Value;
		set => _length.Value = value;
	}

	/// <summary>
	/// WMA period applied to the predicted price.
	/// </summary>
	public int LinearLength
	{
		get => _linearLength.Value;
		set => _linearLength.Value = value;
	}

	/// <summary>
	/// Candle type.
	/// </summary>
	public DataType CandleType
	{
		get => _candleType.Value;
		set => _candleType.Value = value;
	}

	/// <summary>
	/// Constructor.
	/// </summary>
	public LinearCrossTradingStrategy()
	{
		_length = Param(nameof(Length), 21)
			.SetGreaterThanZero()
			.SetDisplay("Length", "Number of candles in the price-on-volume regression", "Indicators");

		_linearLength = Param(nameof(LinearLength), 9)
			.SetGreaterThanZero()
			.SetDisplay("Linear Length", "WMA period applied to the predicted price", "Indicators");

		_candleType = Param(nameof(CandleType), TimeSpan.FromMinutes(5).TimeFrame())
			.SetDisplay("Candle Type", "Type of candles to use", "General");
	}

	/// <inheritdoc />
	public override IEnumerable<(Security sec, DataType dt)> GetWorkingSecurities()
	{
		return [(Security, CandleType)];
	}

	/// <inheritdoc />
	protected override void OnReseted()
	{
		base.OnReseted();
		ResetState();
	}

	private void ResetState()
	{
		_window.Clear();
		_prevPredicted = null;
		_prevWma = null;
		_prevMacd = null;
		_prevLow = null;
	}

	/// <inheritdoc />
	protected override void OnStarted2(DateTime time)
	{
		base.OnStarted2(time);

		ResetState();

		_wma = new WeightedMovingAverage { Length = LinearLength };
		var macd = new MovingAverageConvergenceDivergenceSignal();

		var subscription = SubscribeCandles(CandleType);
		subscription
			.BindEx(macd, ProcessCandle)
			.Start();

		var area = CreateChartArea();
		if (area != null)
		{
			DrawCandles(area, subscription);
			DrawIndicator(area, _wma);
			DrawOwnTrades(area);

			var oscillators = CreateChartArea();
			if (oscillators != null)
				DrawIndicator(oscillators, macd);
		}
	}

	private void ProcessCandle(ICandleMessage candle, IIndicatorValue macdValue)
	{
		if (candle.State != CandleStates.Finished)
			return;

		var prevLow = _prevLow;
		_prevLow = candle.LowPrice;

		_window.Enqueue((candle.ClosePrice, candle.TotalVolume));
		while (_window.Count > Length)
			_window.Dequeue();

		decimal? predicted = null;
		decimal? wma = null;

		if (_window.Count == Length)
		{
			predicted = Predict(candle.TotalVolume);
			var wmaValue = _wma.Process(new DecimalIndicatorValue(_wma, predicted.Value, candle.OpenTime) { IsFinal = true });
			if (_wma.IsFormed)
				wma = wmaValue.GetValue<decimal>();
		}

		var prevPredicted = _prevPredicted;
		var prevWma = _prevWma;
		_prevPredicted = predicted;
		_prevWma = wma;

		if (!macdValue.IsFormed || macdValue is not IMovingAverageConvergenceDivergenceSignalValue { Macd: decimal macd, Signal: decimal signal })
			return;

		var prevMacd = _prevMacd;
		_prevMacd = macd;

		if (prevMacd is not decimal lastMacd || !IsFormedAndOnlineAndAllowTrading())
			return;

		var crossUp = predicted is decimal p && wma is decimal w && prevPredicted is decimal pp && prevWma is decimal pw && pp <= pw && p > w;
		var macdUp = macd > signal && macd > lastMacd;
		var macdDown = macd < signal && macd < lastMacd;
		var lowerLow = prevLow is decimal pl && candle.LowPrice < pl;

		if (crossUp && macdUp && Position <= 0)
			BuyMarket(Volume + Math.Abs(Position));
		else if (macdDown && lowerLow && Position >= 0)
			SellMarket(Volume + Math.Abs(Position));
	}

	private decimal Predict(decimal volume)
	{
		// Least-squares fit close = a + b * volume over the window, evaluated at the current volume.
		var n = _window.Count;
		var meanVolume = _window.Sum(x => x.volume) / n;
		var meanClose = _window.Sum(x => x.close) / n;

		var covariance = 0m;
		var variance = 0m;
		foreach (var (close, vol) in _window)
		{
			var dv = vol - meanVolume;
			covariance += dv * (close - meanClose);
			variance += dv * dv;
		}

		var slope = variance == 0 ? 0 : covariance / variance;
		return meanClose + slope * (volume - meanVolume);
	}
}