Mean Reversion ATR Trend Strategy

This strategy utilizes the highs and lows of price volatility to determine the timing of entries and exits of positions. It aims to establish long positions when price volatility is high and take profits when price…

Pine ScriptMean reversion Trend following Breakout MACD SMA ATR

Author: ChaoZhang
Complexity: Intermediate · 76 lines
Parameters: 7
Backtest bar: 1d · swing
Tested on: BTC/USDT · futures

Jump to code View in vault Original on FMZ

Overview

Strategy Logic

Use ATR indicator to measure price volatility. Calculate the ATR over the last 20 periods and get its moving average and standard deviation. If current ATR value exceeds the average plus one standard deviation, price volatility is considered high.
Use first order logarithmic price change rate to determine price trend. Calculate the logarithmic close price change rate over the last 20 periods, get its moving average. If the current change rate exceeds the average for 3 consecutive days and is positive, price is considered in an uptrend.
When price volatility is high and price shows an uptrend, go long. When price pulls back and stop loss is triggered, close position. Stop loss price is adjusted dynamically to stay below the lowest price minus 2 times ATR.

Advantage Analysis

Utilize price volatility and trend to determine long/short timing, avoid over-trading in ranging markets.
Dynamic stop loss avoids excessive loss from too wide stops.
Backtest shows annualized return of 159% during 2015-2021, far exceeding 120% of buy & hold.

Risk Analysis

Overly aggressive ATR parameters may result in too few entry opportunities. Can relax parameters moderately to increase frequency.
Trend indicator may generate false signals contradicting actual trend. Should add more confirming factors to avoid potential losses.
Backtest period only 6 years. Need larger sample and robustness check to avoid overfitting.
Unable to assess performance in extreme conditions like flash crashes. Manual intervention or stop programming required.

Optimization Directions

Add more trend confirming indicators like MACD, KDJ to improve trend accuracy.
Tune ATR parameters adaptively based on different products and market regimes to optimize volatility gauge.
Add breakout logic and trend accelerating factors to size up on breakouts.
Test different stop loss types like percentage, volatility stop on performance.
Evaluate on metrics like trade frequency, curve stability, max drawdown to ensure robustness.

Summary

This strategy combines the advantages of gauging volatility and trend to determine possible reversal points to enter on amplified volatility, and uses dynamic stops to control risk. Backtest shows decent alpha generated. But 6-year sample is limited, key parameters need market-specific tuning, and more confirming factors are needed to reduce false signals. Comprehensive robustness check also required before applying to live trading. Overall this provides an idea of mean reversion on volatility but still needs refinement and rigorous verification to become a robust quant strategy.

Parameters

Name	Default	What it controls
v_input_1	timestamp(01 Apr 2000 13:30 +0000)	Backtest Start Time
v_input_2	false	Define backtest end-time (If false, will test up to most recent candle)
v_input_3	timestamp(01 May 2021 19:30 +0000)	Backtest End Time (if checked above)
v_input_4	14	Length of ATR for trailing stop loss
v_input_5	2	ATR Multiplier for trailing stop loss
v_input_6	20	Length of ATR to determine volatility
v_input_7	20	Length of Drift

Source Pine Script

/*backtest
start: 2022-09-14 00:00:00
end: 2023-09-20 00:00:00
period: 1d
basePeriod: 1h
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/

// This source code is subject to the terms of the Mozilla Public License 2.0 at https://mozilla.org/MPL/2.0/
// © DojiEmoji (kevinhhl)

//@version=4
strategy("Mean Reversion (ATR) Strategy [KL]",overlay=true,pyramiding=1)
ENUM_LONG = "Long"

// Timeframe {
backtest_timeframe_start = input(defval = timestamp("01 Apr 2000 13:30 +0000"), title = "Backtest Start Time", type = input.time)
USE_ENDTIME = input(false,title="Define backtest end-time (If false, will test up to most recent candle)")
backtest_timeframe_end = input(defval = timestamp("01 May 2021 19:30 +0000"), title = "Backtest End Time (if checked above)", type = input.time)
within_timeframe = true
// }

// Trailing stop loss {
ATR_X2_TSL = atr(input(14,title="Length of ATR for trailing stop loss")) * input(2.0,title="ATR Multiplier for trailing stop loss",type=input.float)
TSL_source = low
var stop_loss_price = float(0)
TSL_line_color = color.green, TSL_transp = 100
if strategy.position_size == 0 or not within_timeframe
    TSL_line_color := color.black
    stop_loss_price := TSL_source - ATR_X2_TSL 
else if strategy.position_size > 0
    stop_loss_price := max(stop_loss_price, TSL_source - ATR_X2_TSL)
    TSL_transp := 0
plot(stop_loss_price, color=color.new(TSL_line_color, TSL_transp))
// }

// Variables for confirmations of entry {
_len_volat = input(20,title="Length of ATR to determine volatility")
_ATR_volat = atr(_len_volat)
_avg_atr = sma(_ATR_volat, _len_volat)
_std_volat = stdev(_ATR_volat,_len_volat)
signal_diverted_ATR = _ATR_volat > (_avg_atr + _std_volat) or _ATR_volat < (_avg_atr - _std_volat)

_len_drift = input(20,title="Length of Drift")//default set to const: _len_vol's default value
_prcntge_chng = log(close/close[1])
_drift = sma(_prcntge_chng, _len_drift) - pow(stdev(_prcntge_chng, _len_drift),2)*0.5
_chg_drift = _drift/_drift[1]-1
signal_uptrend = (_drift > _drift[1] and _drift > _drift[2]) or _drift > 0

entry_signal_all = signal_diverted_ATR and signal_uptrend
// }

alert_per_bar(msg)=>
    prefix = "[" + syminfo.root + "] "
    suffix = "(P=" + tostring(close) + "; atr=" + tostring(_ATR_volat) + ")"
    alert(tostring(prefix) + tostring(msg) + tostring(suffix), alert.freq_once_per_bar)

// MAIN {
if within_timeframe

    if strategy.position_size > 0 and strategy.position_size[1] > 0 and (stop_loss_price/stop_loss_price[1]-1) > 0.005
        alert_per_bar("TSL raised to " + tostring(stop_loss_price))

    // EXIT:
	if strategy.position_size > 0 and TSL_source <= stop_loss_price
	    exit_msg = close <= strategy.position_avg_price ? "stop loss" : "take profit"
        strategy.close(ENUM_LONG, comment=exit_msg)
    // ENTRY:
    else if entry_signal_all and (strategy.position_size == 0 or (strategy.position_size > 0 and close > stop_loss_price))
		entry_msg = strategy.position_size > 0 ? "adding" : "initial"
		strategy.entry(ENUM_LONG, strategy.long, comment=entry_msg)

if strategy.position_size == 0
    stop_loss_price := float(0)
// }

Published by ChaoZhang on the FMZ strategy square · last modified 2023-09-21. 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/Mean-Reversion-ATR-Trend-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.