MACD Momentum Indicator Backtest Strategy

Pine ScriptMomentum Machine learning Mean reversion RSI MACD EMA SMA

Author: ChaoZhang
Complexity: Intermediate · 87 lines
Parameters: 15
Backtest bar: 1h · intraday
Tested on: BTC/USDT · futures

Jump to code View in vault Original on FMZ

Overview

This strategy combines the MACD momentum indicator with the RSI overbought/oversold indicator. When MACD crosses up or down, it checks if RSI also completes the corresponding bottoming/topping reversal over the lookback period to generate more reliable trading signals. Typical short-term mean reversion strategy logic.

Strategy Logic

Calculate MACD DIFF, DEA and histogram. Crossover of DIFF above DEA gives bullish crossover signal, and crossover below gives death cross signal.
Calculate RSI to identify oversold bounces and overbought selloffs. Lookback window checks if recent bottoming or topping has occurred.
When MACD bullish crossover happens, if RSI has bounced off oversold within lookback window, long signal is generated. On MACD death cross, short signal is generated if RSI topped over lookback window.
Stop loss set after entry to control risk.

Advantages

MACD sensitively identifies trend changes. RSI effectively judges overbought/oversold levels.
Requiring both MACD and RSI signals filters out false signals.
Lookback window improves signal reliability.
Stop loss aids risk management.

Risks

Lagging of MACD and RSI may cause missed optimal entries.
Lower probability of dual-indicator signal means fewer trades.
No consideration of larger trend direction risks being trapped.
Poor stop loss tuning may be too wide or too tight.

Possible Solutions:

Adjust MACD and RSI parameters to reduce lag.
Widen indicator threshold ranges to provide more signals.
Add trend filter to avoid counter-trend entries.
Test different stop loss parameters for optimal levels.

Optimization Directions

Test SMA and other moving averages.
Add trailing stop loss for flexible stops.
Incorporate trend strength to judge entry quality.
Use machine learning to predict indicator movements.
Combine more factors to optimize entry timing.

Summary

This strategy filters for reliable reversal signals using coordinated MACD and RSI. The logic is clear and parameters flexible for enhancements like indicator selection, trend filters, stop loss techniques etc to acquire more trades while maintaining stability, but over-optimization risks need to be avoided.

Parameters

Name	Default	What it controls
v_input_1	timestamp(13 Jun 2022)	Start Date
v_input_2	timestamp(13 Jun 2024)	Start Date
v_input_3_close	0	(?RSI Settings)RSI Source: close \| high \| low \| open \| hl2 \| hlc3 \| hlcc4 \| ohlc4
v_input_4	14	Length
v_input_5	30	Over Sold Threshold
v_input_6	70	Over Bought Threshold
v_input_7	7	RSI cross lookback period
v_input_8_close	0	(?MACD Settings)MACD Source: close \| high \| low \| open \| hl2 \| hlc3 \| hlcc4 \| ohlc4
v_input_9	12	Fast Length
v_input_10	26	Slow Length
v_input_int_1	9	Signal Smoothing
v_input_string_1	0	Oscillator MA Type: EMA \| SMA
v_input_string_2	0	Signal Line MA Type: EMA \| SMA
v_input_11	15	(?Stop Loss Settings)Long Stop Loss (%)
v_input_12	15	Short Stop Loss (%)

Source Pine Script

/*backtest
start: 2023-08-24 00:00:00
end: 2023-09-23 00:00:00
period: 1h
basePeriod: 15m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/

//@version=5
//based on Range Strat - MACD/RSI 
// strategy("MACD/RSI - edited", 
//      overlay=true,
//      default_qty_type=strategy.percent_of_equity,
//      default_qty_value=10, precision=2, initial_capital=100000,
//      pyramiding=2,
//      commission_value=0.05)

//Backtest date range
StartDate = input(timestamp("13 Jun 2022"), title="Start Date")
EndDate = input(timestamp("13 Jun 2024"), title="Start Date")
inDateRange = true

// RSI Input Settings
rsisrc = input(title="RSI Source", defval=close, group="RSI Settings")
length = input(title="Length", defval=14, group="RSI Settings" )
overSold = input(title="Over Sold Threshold", defval=30, group="RSI Settings" )
overBought = input(title="Over Bought Threshold", defval=70, group="RSI Settings" )
rsi_lookback = input(title="RSI cross lookback period", defval=7, group="RSI Settings")

// Calculating RSI
vrsi = ta.rsi(rsisrc, length)
co = ta.crossover(vrsi, overSold)
cu = ta.crossunder(vrsi, overBought)

// Function looking for a happened condition during lookback period
f_somethingHappened(_cond, _lookback) =>
    bool _crossed = false
    for i = 1 to _lookback
        if _cond[i]
            _crossed := true
    _crossed


coCheck = f_somethingHappened(co, rsi_lookback)
cuCheck = f_somethingHappened(cu, rsi_lookback)

// MACD Input Settings
macdsrc = input(title="MACD Source", defval=close, group="MACD Settings")
fast_length = input(title="Fast Length", defval=12, group="MACD Settings")
slow_length = input(title="Slow Length", defval=26, group="MACD Settings")
signal_length = input.int(title="Signal Smoothing",  minval = 1, maxval = 50, defval = 9, group="MACD Settings")
sma_source = input.string(title="Oscillator MA Type",  defval="EMA", options=["SMA", "EMA"], group="MACD Settings")
sma_signal = input.string(title="Signal Line MA Type", defval="EMA", options=["SMA", "EMA"], group="MACD Settings")


// Calculating MACD
fast_ma = sma_source == "SMA" ? ta.sma(macdsrc, fast_length) : ta.ema(macdsrc, fast_length)
slow_ma = sma_source == "SMA" ? ta.sma(macdsrc, slow_length) : ta.ema(macdsrc, slow_length)
macd = fast_ma - slow_ma
signal = sma_signal == "SMA" ? ta.sma(macd, signal_length) : ta.ema(macd, signal_length)
delta = macd - signal

MACDcrossover = ta.crossover(delta, 0)
MACDcrossunder = ta.crossunder(delta, 0)

// Stop Loss Input Settings
longLossPerc = input(title="Long Stop Loss (%)", defval=15, group="Stop Loss Settings") * 0.01
shortLossPerc = input(title="Short Stop Loss (%)", defval=15, group="Stop Loss Settings") * 0.01

// Calculating Stop Loss
longStopPrice  = strategy.position_avg_price * (1 - longLossPerc)
shortStopPrice = strategy.position_avg_price * (1 + shortLossPerc)



// Strategy Entry
if (not na(vrsi))
	if (inDateRange and MACDcrossover and coCheck)
		strategy.entry("LONG", strategy.long, comment="LONG")
	if (inDateRange and MACDcrossunder and cuCheck)
		strategy.entry("SHORT", strategy.short, comment="SHORT")

// Submit exit orders based on calculated stop loss price
if (strategy.position_size > 0)
    strategy.exit(id="LONG STOP", stop=longStopPrice)
if (strategy.position_size < 0)
    strategy.exit(id="SHORT STOP", stop=shortStopPrice)

Published by ChaoZhang on the FMZ strategy square · last modified 2023-09-24. 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-Momentum-Indicator-Backtest-Strategy.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.