MACD Trend Following Trading Strategy Based on MACD Indicator

This strategy identifies trend direction based on MACD indicator and makes specific buy and sell decisions using Stoch indicator. It adopts long cycle MACD to determine major trend and short cycle Stoch for entries…

Pine ScriptTrend following Machine learning Mean reversion MACD EMA SMA Stochastic

Author: ChaoZhang
Complexity: Intermediate · 126 lines
Parameters: 26
Backtest bar: 10m · intraday
Tested on: BTC/USDT · futures

Jump to code View in vault Original on FMZ

Overview

Strategy Logic

Judging major trend direction using MACD indicator
- Calculating fast EMA, slow EMA and MACD histogram
- Comparing MACD movements in different cycles to determine trend
Identifying specific buy and sell points using Stoch indicator
- Calculating %K line and %D line
- Divergence near overbought or oversold zone, with Stoch bouncing back as trading signals
Making buy and sell decisions based on trend direction and Stoch signals
- Go long when major cycle MACD rises and Stoch buy signal appears
- Go short when major cycle MACD falls and Stoch sell signal appears
Setting stop loss and take profit to optimize risk management

Advantage Analysis

Combining trend following and overbought-oversold indicator helps capture mid-to-long term trends effectively
MACD determines major direction while Stoch works out trading details, which reduces risks
Making full use of indicator combinations to form systematic strategies
Setting stop loss and take profit controls trading risks
Optimizable parameters adapt to different market conditions

Risk Analysis

Inaccurate mid-to-long term trend judgment may lead to contrarian trading losses
False signals from Stoch cause insufficient profit or losses
Stop loss point may be broken when trend changes, expanding losses
Improper profit target levels affect strategy performance
Ineffective parameters and failure to adapt to changing environment can invalidate the strategy
Risks can be reduced by optimizing trend judgment, verifying Stoch signals, adjusting stop loss and take profit, etc.

Optimization Directions

Optimize MACD parameter mix to improve trend judgment accuracy
Consider multi-cycle Stoch to avoid false signals
Dynamically adjust stop loss and take profit ratios to adapt market volatility
Add other indicator signals to verify and improve validity
Optimize parameters based on different products' characteristics and trading sessions
Introduce machine learning models to assist trend direction judgment
Incorporate volume indicators to avoid insufficient chasing or excessive follow

Conclusion

This strategy integrates the strengths of MACD and Stoch indicators to capture mid-to-long term trends while controlling risks. It can be effective in various market conditions by optimizing parameters, setting stop loss and take profit, verifying signals, etc. There is still room for improvement by further parameter tuning, improving signal accuracy, and incorporating machine learning. The strategy can be more comprehensive and intelligent.

Parameters

Name	Default	What it controls
v_input_1	3	%D Smoothing
v_input_2	14	%K Length
v_input_3	42	%K2 Length
v_input_4	126	%K3 Length
v_input_5	378	%K4 Length
v_input_6	14	%K5 Length
v_input_7	30	%K6 Length
v_input_8	true	%K Smoothing
v_input_9	0.3	buffer
v_input_10	144	Fast Length
v_input_11	312	Slow Length
v_input_12_close	0	Source: close \| high \| low \| open \| hl2 \| hlc3 \| hlcc4 \| ohlc4
v_input_13	108	Signal Smoothing
v_input_14	0	Oscillator MA Type: EMA \| SMA
v_input_15	0	Signal Line MA Type: EMA \| SMA
v_input_16	30	MACDCHAStep Size
v_input_17	20	MACDCHAStep Size2
v_input_18	10	MACDCHAStep Size3
v_input_19	5	MACDCHAStep Size4
v_input_20	3	MACDCHAStep Size5
v_input_21	true	MACDCHAStep Size6
v_input_22	50	histStep Size
v_input_23	4	Long stop loss %
v_input_24	10	Go long and take profit %
v_input_25	4	Short stop loss %
v_input_26	10	Short selling and take profit %

Source Pine Script

/*backtest
start: 2023-09-19 00:00:00
end: 2023-09-26 00:00:00
period: 10m
basePeriod: 1m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/

//@version=4

// strategy(title="Self-use strategyv0.2",calc_on_order_fills=false,calc_on_every_tick =false, initial_capital=10000,commission_type=strategy.commission.percent, commission_value=0.00,overlay = true,default_qty_type = strategy.cash, default_qty_value = 10000)



//STOCH
periodD = input(3, title="%D Smoothing", minval=1)
periodK = input(14, title="%K Length", minval=1)
periodK2 = input(42, title="%K2 Length", minval=1)
periodK3 = input(126, title="%K3 Length", minval=1)
periodK4 = input(378, title="%K4 Length", minval=1)
periodK5 = input(14, title="%K5 Length", minval=1)
periodK6 = input(30, title="%K6 Length", minval=1)
smoothK = input(1, title="%K Smoothing", minval=1)
k = sma(stoch(close, high, low, periodK), smoothK)
k2 = sma(stoch(close, high, low, periodK2), smoothK*3)
k3 = sma(stoch(close, high, low, periodK3), smoothK*3*3)
k4 = sma(stoch(close, high, low, periodK4), smoothK*3*3*3)
d = sma(k, periodD)
all = (k+k2*3+k3*9+k4*18)/31
allp = sma(all, periodK6)



buffer = input(title="buffer", type=input.float, defval=0.3, minval = 0, step = 0.1)
b1 = close[1]* (1+buffer/100)
b2 = close[1]* (1-buffer/100)

//MACD
fast_length = input(title="Fast Length", defval=144)
slow_length = input(title="Slow Length", defval=312)

src = input(title="Source", defval=close)
signal_length = input(title="Signal Smoothing",  minval = 1, maxval = 200, defval = 108)
sma_source = input(title="Oscillator MA Type",  defval="EMA", options=["SMA", "EMA"])
sma_signal = input(title="Signal Line MA Type", defval="EMA", options=["SMA", "EMA"])

// Calculating
fast_ma = sma_source == "SMA" ? sma(src, fast_length) : ema(src, fast_length)
slow_ma = sma_source == "SMA" ? sma(src, slow_length) : ema(src, slow_length)
macd = fast_ma - slow_ma
signal = sma_signal == "SMA" ? sma(macd, signal_length) : ema(macd, signal_length)
hist = macd - signal
MACDCHA = input(title="MACDCHAStep Size", defval=30)
MACDCHA2 = input(title="MACDCHAStep Size2", defval=20)
MACDCHA3 = input(title="MACDCHAStep Size3", defval=10)
MACDCHA4 = input(title="MACDCHAStep Size4", defval=5)
MACDCHA5 = input(title="MACDCHAStep Size5", defval=3)
MACDCHA6 = input(title="MACDCHAStep Size6", defval=1)
HISTCHA = input(title="histStep Size", defval=50)
macdcha = hist - hist[MACDCHA]
macdcha2 = hist - hist[MACDCHA2]
macdcha3 = hist - hist[MACDCHA3]
macdcha4 = hist - hist[MACDCHA4]
macdcha5 = hist - hist[MACDCHA5]
macdcha6 = hist - hist[MACDCHA6]
histcha = hist[HISTCHA]
var true2 = 0
var true2_1 = 0
var true2_2 = 0
var true2_3 = 0
var true2_4 = 0//Extension
var fangxiang =0
//Confirm direction
if(macdcha>=0 and macdcha2>=0 and macdcha3>=0 and macdcha4>=0 and macdcha5>=0 and macdcha6>=0)
    fangxiang := 1
    true2_2 := 0
if(macdcha<=0 and macdcha2<=0 and macdcha3<=0 and macdcha4<=0 and macdcha5<=0 and macdcha6<=0)
    fangxiang :=-1
    true2_1 := 1
//k3min = min(k3,k3[1],k3[2],k3[3],k3[4],k3[5],k3[6],k3[7],k3[8],k3[9],k3[10],k3[11],k3[12],k3[13],k3[14],k3[15],k3[16],k3[17],k3[18],k3[19],k3[20],k3[21],k3[22],k3[23],k3[24],k3[25],k3[26],k3[27],k3[28],k3[29],k3[30],k3[31],k3[32],k3[33],k3[34],k3[35],k3[36],k3[37],k3[38],k3[39],k3[40],k3[41],k3[42],k3[43],k3[44],k3[45],k3[46],k3[47],k3[48],k3[49],k3[50])
//k3max = max(k3,k3[1],k3[2],k3[3],k3[4],k3[5],k3[6],k3[7],k3[8],k3[9],k3[10],k3[11],k3[12],k3[13],k3[14],k3[15],k3[16],k3[17],k3[18],k3[19],k3[20],k3[21],k3[22],k3[23],k3[24],k3[25],k3[26],k3[27],k3[28],k3[29],k3[30],k3[31],k3[32],k3[33],k3[34],k3[35],k3[36],k3[37],k3[38],k3[39],k3[40],k3[41],k3[42],k3[43],k3[44],k3[45],k3[46],k3[47],k3[48],k3[49],k3[50])
allpmax = max(allp[1],allp[2],allp[3],allp[4],allp[5],allp[6])
allpmin = min(allp[1],allp[2],allp[3],allp[4],allp[5],allp[6])
if(histcha < 0 and macdcha>=0 and macdcha2>=0 and macdcha3>=0 and macdcha4>=0 and macdcha5>=0 and macdcha6>=0 and d < 20 and volume > volume[1] and true2_1 == 1 and allp>allp[1] and allp <80)//and k3max < 80  //and k3min < 30 and k3 >20 and k2<50
    strategy.entry("open long", true, comment = "open long") // and close > close[1] and cci1> MEA1
    true2_1 :=0
if(d >80)
    strategy.close( "open long", comment = "close long")
    true2_1 :=1

stop_loss=input(4, "Long stop loss %", minval = 1, step = 1)
sl = strategy.position_avg_price * (1-stop_loss/100)
close_Stop = close < sl
if(close_Stop or(allp<20 and allp[1]>20))
    strategy.close( "open long", comment = "Long stop loss")
    true2_1 :=1
Target_profit=input(10, "Go long and take profit %", minval = 1, step = 1)
tp = strategy.position_avg_price * (1+Target_profit/100)
close_Target = close > tp
strategy.close("open long", when = close_Target, comment ="Go long and make profit")


//Empty
if(histcha > 0 and macdcha<=0 and macdcha2<=0 and macdcha3<=0 and macdcha4<=0 and macdcha5<=0 and macdcha6<=0 and d > 80 and volume > volume[1] and true2_2 == 1 and allp<allp[1] and allp >20) // and k3max>70 and k3<80
    //strategy.entry("open short", comment = "open short") 
    strategy.entry("open short", strategy.short,comment ="open short")
    true2_2 := 0
if( d <20)
   // strategy.close(  comment = "close short")
    strategy.close("open short",  comment = "close short")
    true2_2 := 1

stop_loss2=input(4, "Short stop loss %", minval = 1, step = 1)
sl2 = strategy.position_avg_price * (1+stop_loss2/100)
close_Stop2 = close > sl2
if(close_Stop2 or(allp>80 and allp[1]<80))
    strategy.close( "open short", comment = "Short stop loss")
    true2_2 == 1
Target_profit2=input(10, "Short selling and take profit %", minval = 1, step = 1)
tp2 = strategy.position_avg_price * (1-Target_profit2/100)
close_Target2 = close < tp2
strategy.close("open short", when = close_Target2, comment ="Short selling profit")

Published by ChaoZhang on the FMZ strategy square · last modified 2023-09-27. 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/MACD-Trend-Following-Trading-Strategy-Based-on-MACD-Indicator.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.