Dynmaic Risk Adjusted Momentum Trading Strategy

Pine ScriptMomentum Trend following Volatility SMA ATR

Author: ChaoZhang
Complexity: Intermediate · 102 lines
Parameters: 17
Backtest bar: 1d · swing
Tested on: BTC/USDT · futures

Jump to code View in vault Original on FMZ

Overview

This unique systematic rulebased trading strategy is in the trend following category. It uses normalised prices series transformed from raw ticker price to gernerate trading signals. Advanced position sizing and risk management techniques, commonly reserved for institutional portfolio management, are utilised in this strategy - proven positioning and risk control technologies used by financial advisers like Commodity Trading Advisors and Managed Futures funds.

Strategy mechanics

The "normalised price" is a volatility-adjusted accumulated daily returns series. Daily volatility adjustment lookback is user defined. Hull moving average of the normalised price is used as the main trend indicator. Lookback period of the HMA is user defined too, with default period of 100 days for a responsive signal without inducing over-trading.

The core trades are simple, long when normalised price crossover HMA, short when crossunder HMA. New signals close any existing opposing position.

Position size is dynamically adjusted based on recent price volatility and the user defined annual risk target. Positions are risk-weighted, larger size with lower volatility and smaller with higher volatility. Recent volatility is the standard deviation of returns over the last 14 periods, then extrapolated into annual volatility as expeted returns. Annual risk target is used as reference for volatility adjusted position sizing. Default target is 10% of total capital. Initial capital should be set as maximum loss per trade. Max leverage allows achieving risk target if underlying natural volatility is insuffient, and alleviates excessively low volatility.

Hard stops are based on recent price average true range multiplier, user configurable.

Advantages

Normalised prices reduces false signals
Dynamic position sizing controls risk effectively
Hard stops prevent runaway losses
Simple trend following logic for transparency

Risks

Lagging issues with Hull moving average
Capping profits while lowering risk through volatility adjusted position sizing
Stops too tight vulnerable to spikes

Risk controls measures include alternate moving average selections, adjusting risk targets.

Enhancements

Test effectiveness of other moving average types
Optimize parameters of moving averages
Try long-only or short-only variants
Fine tune stop loss ranges
Experiment with other stop loss mechanisms

Conclusion

The strategy integrates various techniques like normalisation, dynamic position adjustment, hard stops to control risks. Trading is based on simple trend following rules. Parameters can be adjusted for personal preferences and market regimes. Worth further testing and verification for viable real world application.

Parameters

Name	Default	What it controls
v_input_1_close	0	(?Strategy Settings)Source: close \| high \| low \| open \| hl2 \| hlc3 \| hlcc4 \| ohlc4
v_input_int_1	14	Lookback period for price normalisation filter
v_input_int_2	100	Lookback period for Hull Moving Average
v_input_int_3	false	HMA Offset
v_input_2	true	Long
v_input_3	true	Short
v_input_float_1	true	(?Risk Management Settings)Stop multiple
v_input_float_2	true	Max Leverage
v_input_float_3	10	Annualised Volatility Target %
v_input_4	false	Compounding
v_input_float_4	50	%
v_input_int_4	true	(?Backtest range)From Day
v_input_int_5	true	From Mon
v_input_int_6	2018	From Yr
v_input_int_7	true	To Day
v_input_int_8	true	To Mon
v_input_int_9	9999	To Yr

Source Pine Script

/*backtest
start: 2023-01-17 00:00:00
end: 2024-01-23 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/
// © Crunchster1

//@version=5
strategy(title="Crunchster's Normalised Trend Strategy", shorttitle="Normalised Trend Strategy", overlay=false )

// Inputs and Parameters
src = input(close, 'Source', group='Strategy Settings')
length = input.int(title="Lookback period for price normalisation filter", defval=14, minval=2, group='Strategy Settings', tooltip='This sets the lookback period for the volatility adjustment of returns, which is used to transform the price series into the "real price"')
hlength = input.int(title="Lookback period for Hull Moving Average", defval=100, minval=2, group='Strategy Settings')
offset = input.int(title="HMA Offset", defval=0, minval=0, group='Strategy Settings')
long = input(true, 'Long', inline='08', group='Strategy Settings')
short = input(true, 'Short', inline='08', group='Strategy Settings', tooltip='Toggle long/short strategy on/off')

stopMultiple = input.float(1, 'Stop multiple', step=0.25, group='Risk Management Settings', tooltip='Multiple for ATR, setting hard stop loss from entry price')
lev = input.float(1, 'Max Leverage', step=0.5, group='Risk Management Settings', tooltip='Max leverage sets maximum allowable leverage of total capital (initial capital + any net profit), capping maximum volatility adjusted position size')
riskT = input.float(10, maxval=75, title='Annualised Volatility Target %', group='Risk Management Settings', tooltip='Specify annual risk target, used to determine volatility adjusted position size. Annualised daily volatility is referenced to this value and position size adjusted accordingly')
comp = input(false, 'Compounding', inline='09', group='Risk Management Settings')
Comppct = input.float(50, '%', step=5, inline='09', group='Risk Management Settings', tooltip='Toggle compounding of profit, and set % of profit to compound')

// Backtesting period
FromDay = input.int(defval=1, title='From Day', minval=1, maxval=31, inline='04', group='Backtest range')
FromMonth = input.int(defval=1, title='From Mon', minval=1, maxval=12, inline='04', group='Backtest range')
FromYear = input.int(defval=2018, title='From Yr', minval=1900, inline='04', group='Backtest range', tooltip='Set start of backtesting period')
ToDay = input.int(defval=1, title='To Day', minval=1, maxval=31, inline='05', group='Backtest range')
ToMonth = input.int(defval=1, title='To Mon', minval=1, maxval=12, inline='05', group='Backtest range')
ToYear = input.int(defval=9999, title='To Yr', minval=1900, inline='05', group='Backtest range', tooltip='Set end of backtesting period')

start = timestamp(FromYear, FromMonth, FromDay, 00, 00)
finish = timestamp(ToYear, ToMonth, ToDay, 23, 59)
window = true

// Normalised returns calculation
nRet = (src - src[1]) / ta.stdev((src - src[1]), length)

nPrice = ta.cum(nRet)

//Hull Moving Average - using normalised price series
fHMA = ta.wma(2 * ta.wma(nPrice[offset], hlength / 2) - ta.wma(nPrice[offset], hlength), math.round(math.sqrt(hlength)))

//Risk Management formulae
strategy.initial_capital = 50000
tr = math.max(high - low, math.abs(high - close), math.abs(low - close)) //True range
stopL = ta.sma(tr, 14) //Average true range
stdev = ta.stdev(close-close[1], 14) //volatility of recent returns
maxcapital = strategy.initial_capital+strategy.netprofit //Maximum capital available to invest - initial capital net of profit
annvol = 100*math.sqrt(365)*stdev/close //converts recent volatility of returns into annualised volatility of returns - assumes daily timeframe

risk = 1.1
if comp
    risk := (strategy.initial_capital+(Comppct*strategy.netprofit/100))//adjust investment capital to include compounding
else
    risk := strategy.initial_capital

shares = (risk * (riskT/annvol)) / close //calculates volatility adjusted position size, dependent on user specified annualised risk target
if ((shares*close) > lev*maxcapital) //ensures position size does not exceed available capital multiplied by user specified maximum leverage
    shares := lev*maxcapital/close

//To set the price at the entry point of trade
Posopen() =>
    math.abs(strategy.position_size[1]) <= 0 and math.abs(strategy.position_size) > 0

var float openN = na
if Posopen()
    openN := stopL

// Strategy Rules
if long
    longCondition = ta.crossover(nPrice, fHMA) and window
    exitlong = ta.crossunder(nPrice, fHMA)
    if (longCondition)
        strategy.entry('Go Long!', strategy.long, qty=shares)
    if strategy.position_size > 0    
        strategy.exit('Stop Long', from_entry = 'Go Long!', stop=(strategy.opentrades.entry_price(0) - (openN * stopMultiple)))
    if (exitlong)
        strategy.close('Go Long!', immediately = true)

if short
    shortCondition = ta.crossunder(nPrice, fHMA) and window
    exitshort = ta.crossover(nPrice, fHMA)
    if (shortCondition)
        strategy.entry('Go Short!', strategy.short, qty=shares)
    if strategy.position_size < 0   
        strategy.exit('Stop Short', from_entry = 'Go Short!', stop=(strategy.opentrades.entry_price(0) + (openN * stopMultiple)))
    if (exitshort)
        strategy.close('Go Short!', immediately = true)

// Visuals of trend and direction
plot(nPrice, title='Real Price', color=color.black)

MAColor = fHMA > fHMA[3] ? #00ff00 : #ff0000
MA1 = plot(fHMA, title='Hull MA', color=MAColor)
MA2 = plot(fHMA[3], title='Hull MA Offset', color=MAColor)
fill(MA1, MA2, title='Band Filler', color=MAColor)

Published by ChaoZhang on the FMZ strategy square · last modified 2024-01-24. 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/Dynmaic-Risk-Adjusted-Momentum-Trading-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.