Integrated Multi Strategy Quantitative Trading System

Pine ScriptBreakout Mean reversion Momentum MACD EMA Heikin Ashi

Author: ChaoZhang
Complexity: Intermediate · 138 lines
Parameters: 13
Backtest bar: 45m · intraday
Tested on: BTC/USDT · futures

Jump to code View in vault Original on FMZ

This article will introduce you to a quantitative trading strategy called "Integrated Multi-Strategy Quantitative Trading System" for cryptocurrencies. This strategy integrates the advantages of multiple single strategies to construct a portfolio of multiple strategies, aiming to achieve higher stability and diversity.

This strategy incorporates four common quantitative trading strategies:

Channel breakout strategy: Construct upper and lower channels based on highest and lowest prices over a certain period and take positions when price breaks out of the channel.
Momentum strategy: Determine momentum based on price changes over a certain period, go long when prices rise accelerating, go short when prices fall accelerating.
MACD strategy: Determine long and short positions based on MACD golden cross and dead cross.
Harami pattern strategy: Identify potential future reversals by recognizing specific candlestick patterns and trade around turning points.

These strategies each have advantages. Combined together, they can achieve more stable returns. Specifically:

The channel breakout strategy can capture market trends; the momentum strategy can timely track short-term trends; the MACD strategy can discover medium-term trend reversals; the Harami strategy can determine key reversal points.

Integrating them into one strategy allows you to chase rises and kill falls during trending markets and open reverse positions around inflection points. Meanwhile, different strategies can also achieve risk diversification.

Of course, such multi-strategy combinations also have some disadvantages:

The strategy is too complex and parameters are difficult to adjust.
There may be conflicts between some strategies.
It increases trading frequency and transaction costs.
Backtesting results may be worse than a single strategy.

Therefore, when using this multi-strategy combination, users should pay attention to the difficulty of parameter adjustment, test the interaction between conflicts, control trading frequency, and conduct sufficient backtesting to ensure its long-term stability.

In general, this integrated multi-strategy quantitative trading system can obtain very rich trading combinations and performs very well in major trends. It combines the advantages of different strategies and can achieve more stable long-term positive returns. It is worth users' further research and optimization to develop a powerful quantitative strategy portfolio.

Parameters

Name	Default	What it controls
v_input_1	5	Length
v_input_2	12	length1
v_input_3	12	fastLength
v_input_4	26	slowlength
v_input_5	9	MACDLength
v_input_6	60	Doji, Min % of Range of Candle for Wicks
v_input_7	false	Doji, Previous Candle Min Pip Body Size
v_input_8	true	Show Price Action Bar Names
v_input_9	false	Highlight Harami & Doji Bars
v_input_10	false	Show Only Harami Style Doji's
v_input_11	true	Generate Alert for Harami & Doji Bars
v_input_12	true	Use Heikin Ashi Candles for Calculations
v_input_13	3	Doji, Number of Lookback Bars

Source Pine Script

/*backtest
start: 2023-09-07 00:00:00
end: 2023-09-14 00:00:00
period: 45m
basePeriod: 5m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/

//@version=3

//Channel breakout
strategy("all_strategy", overlay=true)

length = input(title="Length", minval=1, maxval=1000, defval=5)

upBound = highest(high, length)
downBound = lowest(low, length)

if (not na(close[length]))
    strategy.entry("ChBrkLE", strategy.long, stop=upBound + syminfo.mintick, comment="ChBrkLE")
    strategy.entry("ChBrkSE", strategy.short, stop=downBound - syminfo.mintick, comment="ChBrkSE")

//plot(strategy.equity, title="equity", color=red, linewidth=2, style=areabr)


//Momentum
length1 = input(12)
price = close

momentum(seria, length) =>
    mom = seria - seria[length1]
    mom

mom0 = momentum(price, length1)
mom1 = momentum( mom0, 1)

if (mom0 > 0 and mom1 > 0)
    strategy.entry("MomLE", strategy.long, stop=high+syminfo.mintick, comment="MomLE")
else
    strategy.cancel("MomLE")

if (mom0 < 0 and mom1 < 0)
    strategy.entry("MomSE", strategy.short, stop=low-syminfo.mintick, comment="MomSE")
else
    strategy.cancel("MomSE")

//plot(strategy.equity, title="equity", color=red, linewidth=2, style=areabr)


//MACD Strategy
fastLength = input(12)
slowlength = input(26)
MACDLength = input(9)

MACD = ema(close, fastLength) - ema(close, slowlength)
aMACD = ema(MACD, MACDLength)
delta = MACD - aMACD

if (crossover(delta, 0))
    strategy.entry("MacdLE", strategy.long, comment="MacdLE")

if (crossunder(delta, 0))
    strategy.entry("MacdSE", strategy.short, comment="MacdSE")

//plot(strategy.equity, title="equity", color=red, linewidth=2, style=areabr)


//Harami

pctDw = input(60,minval=0,maxval=90,title="Doji, Min % of Range of Candle for Wicks")
pipMin= input(0,minval=0,title="Doji, Previous Candle Min Pip Body Size")
sname=input(true,title="Show Price Action Bar Names")
cbar = input(false,title="Highlight Harami & Doji Bars")
sHm    = input(false,title="Show Only Harami Style Doji's")
setalm = input(true, title="Generate Alert for Harami & Doji Bars")
uha   =input(true, title="Use Heikin Ashi Candles for Calculations")
bars = input(3,minval=1,maxval=3,step=1, title="Doji, Number of Lookback Bars")
//
// Use only Heikinashi Candles for all calculations
srcclose = uha ? security(heikinashi(syminfo.tickerid), timeframe.period, close) : close
srcopen = uha ? security(heikinashi(syminfo.tickerid), timeframe.period, open) : open
srchigh = uha ? security(heikinashi(syminfo.tickerid), timeframe.period, high) : high
srclow = uha ?security(heikinashi(syminfo.tickerid), timeframe.period, low) : low

//
pip = syminfo.mintick
range = srchigh - srclow


// Calculate Doji/Harami Candles
pctCDw = (pctDw/2) * 0.01
pctCDb = (100-pctDw) * 0.01

//Lookback Candles for bulls or bears
lbBull = bars==1? srcopen[1]>srcclose[1]: bars==2? (srcopen[1]>srcclose[1] and srcopen[2]>srcclose[2]): bars==3?(srcopen[1]>srcclose[1] and srcopen[2]>srcclose[2] and srcopen[3]>srcclose[3]):false
lbBear = bars==1? srcopen[1]<srcclose[1]: bars==2? (srcopen[1]<srcclose[1] and srcopen[2]<srcclose[2]): bars==3?(srcopen[1]<srcclose[1] and srcopen[2]<srcclose[2] and srcopen[3]<srcclose[3]):false

//Lookback Candle Size only if mininum size is > 0
lbSize = pipMin==0? true : bars==1 ? (abs(srcopen[1]-srcclose[1])>pipMin*pip) :
  bars==2 ? (abs(srcopen[1]-srcclose[1])>pipMin*pip and abs(srcopen[2]-srcclose[2])>pipMin*pip) :
  bars==3 ? (abs(srcopen[1]-srcclose[1])>pipMin*pip and abs(srcopen[2]-srcclose[2])>pipMin*pip and abs(srcopen[3]-srcclose[3])>pipMin*pip) :
  false

dojiBu = (srcopen[1] >= max(srcclose,srcopen) and srcclose[1]<=min(srcclose,srcopen)) and lbSize and
  (abs(srcclose-srcopen)<range*pctCDb and (srchigh-max(srcclose,srcopen))>(pctCDw*range) and (min(srcclose,srcopen)-srclow)>(pctCDw*range))? 1 : 0

dojiBe = (srcclose[1] >= max(srcclose,srcopen) and srcopen[1]<=min(srcclose,srcopen)) and lbSize and
  (abs(srcclose-srcopen)<range*pctCDb and (srchigh-max(srcclose,srcopen))>(pctCDw*range) and (min(srcclose,srcopen)-srclow)>(pctCDw*range))? 1 : 0
  
haramiBull = (srcopen<=srcclose or (max(srcclose,srcopen)-min(srcclose,srcopen))<pip*0.5) and lbBull and dojiBu
haramiBear = (srcopen>=srcclose or (max(srcclose,srcopen)-min(srcclose,srcopen))<pip*0.5) and lbBear and dojiBe

dojiBull = not sHm and not haramiBull and not haramiBear and lbBull and dojiBu
dojiBear = not sHm and not haramiBull and not haramiBear and lbBear and dojiBe

//
plotshape(haramiBear and sname?srchigh:na,title="Bearish Harami",text='Bearish\nHarami',color=red, style=shape.arrowdown,location=location.abovebar)
plotshape(haramiBear and cbar?max(srcopen,srcclose):na,title="Bear Colour Harami",color=red, style=shape.circle,location=location.absolute,size=size.normal)
//
plotshape(haramiBull and sname?srclow:na,title="Bullish Harami",text='Bullish\nHarami',color=green, style=shape.arrowup,location=location.belowbar)
plotshape(haramiBull and cbar?max(srcopen,srcclose):na,title="Bull Colour Harami",color=green, style=shape.circle,location=location.absolute,size=size.normal)
//
plotshape(dojiBear and sname?srchigh:na,title="Bearish Doji",text='Bearish\nDoji',color=fuchsia, style=shape.arrowdown,location=location.abovebar)
plotshape(dojiBear and cbar?max(srcopen,srcclose):na,title="Bear Colour Doji",color=fuchsia, style=shape.circle,location=location.absolute,size=size.normal)
//
plotshape(dojiBull and sname?srclow:na,title="Bullish Doji",text='Bullish\nDoji',color=aqua, style=shape.arrowup,location=location.belowbar)
plotshape(dojiBull and cbar?max(srcopen,srcclose):na,title="Bull Colour Doji",color=aqua, style=shape.circle,location=location.absolute,size=size.normal)

// Only Alert harami Doji's
bcolor = haramiBull ? 1 : haramiBear ? 2 : dojiBull ? 3 : dojiBear ? 4 : 0
baralert = setalm and bcolor>0
alertcondition(baralert,title="PACDOJI Alert",message="PACDOJI Alert")

//
plotshape(na(baralert[1])?na:baralert[1], transp=0,style=shape.circle,location=location.bottom, offset=-1,title="Bar Alert Confirmed", 
  color=bcolor[1]==1 ? green : bcolor[1]==2? red : bcolor[1]==3? aqua : bcolor[1]==4? fuchsia : na)

//EOF

Published by ChaoZhang on the FMZ strategy square · last modified 2023-09-15. Collected in the Quant Trading Vault. Unvalidated research: re-test with your own data and costs before trusting it.

Build this with AgenKit

A spec is not a trading system. Paste this into your coding agent and AgenKit runs a specialist team — architecture, test-first build, risk gates, review — with paper trading on by default.

in Claude Code, Cursor, Codex & 4 more

/agenkit build a production-grade trading bot with a backtest engine, risk controls and paper trading by default from strategies/Integrated-Multi-Strategy-Quantitative-Trading-System.md

npx agenkit install engineering-kit in your project
Paste the command into your agent and approve each phase
Get tests, backtest, risk controls and a paper-trading switch

Get AgenKit How /agenkit works

Understand it first

Before you run this, read the lesson behind the idea and how to backtest it honestly. Specs like this one are the survivors of an unknown number of trials; deflate their results accordingly.

Parameters

Source Pine Script

Build this with AgenKit

Understand it first

Related specs