Quantized Gradual Weighted DCA Trading Strategy

Author: ChaoZhang, Date: 2023-11-16 11:32:12
Tags:

img

Overview

The Quantized Gradual Weighted DCA Trading Strategy is a quantitative trading strategy that combines moving average indicators for signal triggering and gradual weighted dollar cost averaging mechanisms. The strategy aims to achieve relatively stable returns in strongly trending markets through trend identification and cost averaging.

Principles

The strategy consists of three main components:

  1. Entry signal judgement

    It uses fast and slow moving average crossovers as the entry signal. Users can choose between SMA, EMA or HMA for the fast and slow moving averages. When the fast MA crosses above the slow MA, a buy signal is generated. When the fast MA crosses below the slow MA, a sell signal is generated.

  2. Gradual weighted DCA

    After a buy signal, the strategy will immediately open a base position. As price continues to fall, the strategy will gradually increase the size of additional safety positions in a weighted manner. The price of each new safety position will be lowered by a fixed percentage relative to the previous one. Also, the allocated funds for each new safety position will be amplified by a factor.

    This gradual increase in position size allows a form of cost averaging, obtaining a better average cost while keeping risks in control.

  3. Take profit and Stop loss

    When price rises above the take profit line, the strategy will close positions for profit. When price falls below the stop loss line, the strategy will exit positions to control loss.

    The take profit line is fixed at base position average price * (1 + fixed percentage).

    The stop loss line fluctuates based on the last safety position price, fixed percentage below it.

Advantages

  1. Combining trend and cost averaging makes it more stable

    Using trends avoids meaningless whipsaws, and cost averaging provides better entry costs.

  2. Gradual position sizing controls risk

    Fixed amplification of safety position sizes, with re-entry threshold, keeps risk in check.

  3. Real-time used funds monitoring

    Incorporated balance usage indicator prevents over-leveraging and forced liquidations.

  4. Separate TP/SL for each position

    Independent exits allow securing profits and cutting losses.

Risks and Improvements

  1. Price whipsaws can trigger multiple safety orders

    In extreme volatility, multiple unnecessary safety orders may be added, increasing loss. Can optimize safety order re-entry threshold.

  2. Moving average parameters need optimization

    Different instruments need different moving average periods. Parameter tuning required.

  3. TP/SL levels need backtest optimization

    TP/SL ratios determine risk/reward profile. Optimal levels vary.

  4. Add maximum drawdown or holding time based forced exit

    Can test incorporating forced exits based on drawdown or holding time to further limit risks.

Summary

The Quantized Gradual Weighted DCA Trading Strategy combines the advantages of trend trading and cost averaging to produce steady returns in strong trends. With optimized parameters, position sizing and re-entry thresholds, it can achieve stable trades with controlled risk. Applicable for hedge funds, CTAs and market neutral strategies.


/*backtest
start: 2022-11-09 00:00:00
end: 2023-11-15 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/
// © MGTG

//@version=5
Strategy = input.string('Long', options=['Long'], group='Strategy', inline='1',
 tooltip='Long bots profit when asset prices rise, Short bots profit when asset prices fall'
 + '\n\n' + 'Please note: to run a Short bot on a spot exchange account, you need to own the asset you want to trade. The bot will sell the asset at the current chart price and buy it back at a lower price - the profit made is actually trapped equity released from an asset you own that is declining in value.')

Profit_currency = input.string('Quote (USDT)', 'Profit currency', options=['Quote (USDT)', 'Quote (BTC)', 'Quote (BUSD)'], group='Strategy', inline='1')
Base_order_size = input.int(10, 'Base order Size', group='Strategy', inline='2', 
 tooltip='The Base Order is the first order the bot will create when starting a new deal.')
Safety_order_size = input.int(20, 'Safety order Size', group='Strategy', inline='2',
 tooltip="Enter the amount of funds your Safety Orders will use to Average the cost of the asset being traded, this can help your bot to close deals faster with more profit. Safety Orders are also known as Dollar Cost Averaging and help when prices moves in the opposite direction to your bot's take profit target.")

Triger_Type = input.string('Over', 'Entry at Cross Over / Under', options=['Over', 'Under'], group='Deal start condition > Trading View custom signal', inline='1',
 tooltip='Deal start condition decision')

Short_Moving_Average  = input.string('SMA', 'Short Moving Average', group='Deal start condition > Trading View custom signal', inline='2',
 options=["SMA", "EMA", "HMA"])
Short_Period         = input.int(5, 'Period', group='Deal start condition > Trading View custom signal', inline='2')
Long_Moving_Average  = input.string('HMA', 'Long Moving Average', group='Deal start condition > Trading View custom signal', inline='3',
 options=["SMA", "EMA", "HMA"])

Long_Period          = input.int(50, 'Period', group='Deal start condition > Trading View custom signal', inline='3')

Target_profit = input.float(1.5, 'Target profit (%)', step=0.05, group='Take profit / Stop Loss', inline='1') * 0.01
Stop_Loss = input.int(15, 'Stop Loss (%)', group='Take profit / Stop Loss', inline='1',
 tooltip='This is the percentage that price needs to move in the opposite direction to your take profit target, at which point the bot will execute a Market Order on the exchange account to close the deal for a smaller loss than keeping the deal open.'
 + '\n' + 'Please note, the Stop Loss is calculated from the price the Safety Order at on the exchange account and not the Dollar Cost Average price.') * 0.01

Max_safety_trades_count = input.int(10, 'Max safety trades count', maxval=10, group='Safety orders', inline='1')
Price_deviation = input.float(0.4, 'Price deviation to open safety orders (% from initial order)', step=0.01, group='Safety orders', inline='2') * 0.01
Safety_order_volume_scale = input.float(1.8, 'Safety order volume scale', step=0.01, group='Safety orders', inline='3')
Safety_order_step_scale = input.float(1.19, 'Safety order step scale', step=0.01, group='Safety orders', inline='3')

// daily_volume  = input.int(500, "Don't start deal(s) if the daily volume is less than", group='Advanced settings', inline='1')
// Minimum_price  = input.int(500, "Minimum price to open deal", group='Advanced settings', inline='1')
// Maximum_price  = input.int(500, "Maximum price to open deal", group='Advanced settings', inline='1')

// Close_deal_after_timeout  = input.int(5, "Close deal after timeout (Hrs)", group='Advanced settings', inline='1')

initial_capital = 8913

strategy(
 title='3Commas Visible DCA Strategy', 
 overlay=true, 
 initial_capital=initial_capital, 
 pyramiding=11, 
 process_orders_on_close=true, 
 commission_type=strategy.commission.percent, 
 commission_value=0.01, 
 max_bars_back=5000, 
 max_labels_count=50)


// Position
status_none  = strategy.position_size == 0
status_long  = strategy.position_size[1] == 0 and strategy.position_size > 0
status_long_offset  = strategy.position_size[2] == 0 and strategy.position_size[1] > 0
status_short = strategy.position_size[1] == 0 and strategy.position_size < 0
status_increase = strategy.opentrades[1] < strategy.opentrades

Short_Moving_Average_Line = 
 Short_Moving_Average == 'SMA' ? ta.sma(close, Short_Period) :
 Short_Moving_Average == 'EMA' ? ta.ema(close, Short_Period) :
 Short_Moving_Average == 'HMA' ? ta.sma(close, Short_Period) : na

Long_Moving_Average_Line = 
 Long_Moving_Average == 'SMA' ? ta.sma(close, Long_Period) :
 Long_Moving_Average == 'EMA' ? ta.ema(close, Long_Period) :
 Long_Moving_Average == 'HMA' ? ta.sma(close, Long_Period) : na
 
Base_order_Condition      = Triger_Type == "Over" ? ta.crossover(Short_Moving_Average_Line, Long_Moving_Average_Line) : ta.crossunder(Short_Moving_Average_Line, Long_Moving_Average_Line) // Buy when close crossing lower band

safety_order_deviation(index) => Price_deviation * math.pow(Safety_order_step_scale,  index - 1)

pd = Price_deviation
ss = Safety_order_step_scale

step(i) =>
 i == 1 ? pd :
 i == 2 ? pd + pd * ss :
 i == 3 ? pd + (pd + pd * ss) * ss :
 i == 4 ? pd + (pd + (pd + pd * ss) * ss) * ss : 
 i == 5 ? pd + (pd + (pd + (pd + pd * ss) * ss) * ss) * ss : 
 i == 6 ? pd + (pd + (pd + (pd + (pd + pd * ss) * ss) * ss) * ss) * ss : 
 i == 7 ? pd + (pd + (pd + (pd + (pd + (pd + pd * ss) * ss) * ss) * ss) * ss) * ss : 
 i == 8 ? pd + (pd + (pd + (pd + (pd + (pd + (pd + pd * ss) * ss) * ss) * ss) * ss) * ss) * ss : 
 i == 9 ? pd + (pd + (pd + (pd + (pd + (pd + (pd + (pd + pd * ss) * ss) * ss) * ss) * ss) * ss) * ss) * ss : 
 i == 10 ? pd + (pd + (pd + (pd + (pd + (pd + (pd + (pd + (pd + pd * ss) * ss) * ss) * ss) * ss) * ss) * ss) * ss) * ss : na

long_line(i) =>
 close[1] - close[1] * (step(i))


Safe_order_line(i) =>
 i == 0 ? ta.valuewhen(status_long, long_line(0), 0) :
 i == 1 ? ta.valuewhen(status_long, long_line(1), 0) :
 i == 2 ? ta.valuewhen(status_long, long_line(2), 0) :
 i == 3 ? ta.valuewhen(status_long, long_line(3), 0) :
 i == 4 ? ta.valuewhen(status_long, long_line(4), 0) :
 i == 5 ? ta.valuewhen(status_long, long_line(5), 0) :
 i == 6 ? ta.valuewhen(status_long, long_line(6), 0) :
 i == 7 ? ta.valuewhen(status_long, long_line(7), 0) :
 i == 8 ? ta.valuewhen(status_long, long_line(8), 0) : 
 i == 9 ? ta.valuewhen(status_long, long_line(9), 0) :
 i == 10 ? ta.valuewhen(status_long, long_line(10), 0) : na

TP_line = strategy.position_avg_price * (1 + Target_profit) 

SL_line = Safe_order_line(Max_safety_trades_count) * (1 - Stop_Loss)

safety_order_size(i) => Safety_order_size * math.pow(Safety_order_volume_scale, i - 1)


plot(Short_Moving_Average_Line, 'Short MA', color=color.new(color.white, 0), style=plot.style_line)
plot(Long_Moving_Average_Line, 'Long MA', color=color.new(color.green, 0), style=plot.style_line)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 1 ? Safe_order_line(1) : na, 'Safety order1', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 2 ? Safe_order_line(2) : na, 'Safety order2', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 3 ? Safe_order_line(3) : na, 'Safety order3', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 4 ? Safe_order_line(4) : na, 'Safety order4', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 5 ? Safe_order_line(5) : na, 'Safety order5', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 6 ? Safe_order_line(6) : na, 'Safety order6', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 7 ? Safe_order_line(7) : na, 'Safety order7', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 8 ? Safe_order_line(8) : na, 'Safety order8', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 9 ? Safe_order_line(9) : na, 'Safety order9', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 and Max_safety_trades_count >= 10 ? Safe_order_line(10) : na, 'Safety order10', color=color.new(#009688, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 ? TP_line : na, 'Take Profit', color=color.new(color.orange, 0), style=plot.style_linebr)
plot(strategy.position_size > 0 ? SL_line : na, 'Safety', color=color.new(color.aqua, 0), style=plot.style_linebr)


currency = 
 Profit_currency == 'Quote (USDT)' ? ' USDT' :
 Profit_currency == 'Quote (BTC)'  ? ' BTC' :
 Profit_currency == 'Quote (BUSD)' ? ' BUSD' : na
 

if Base_order_Condition
    strategy.entry('Base order', strategy.long, qty=Base_order_size/close, when=Base_order_Condition and strategy.opentrades == 0,
     comment='BO' + ' - ' + str.tostring(Base_order_size) + str.tostring(currency))

for i = 1 to Max_safety_trades_count by 1
    i_s = str.tostring(i)
    strategy.entry('Safety order' + i_s, strategy.long, qty=safety_order_size(i)/close,
     limit=Safe_order_line(i), when=(strategy.opentrades <= i) and strategy.position_size > 0, 
     comment='SO' + i_s + ' - ' + str.tostring(safety_order_size(i))  + str.tostring(currency))


for i = 1 to Max_safety_trades_count by 1
    i_s = str.tostring(i)
    // strategy.close('Base order', when=shortCondition)
    // strategy.close('Safety order' + i_s, when=shortCondition)
    // strategy.cancel('Safety order' + i_s, when=shortCondition)
    strategy.cancel('SO' + i_s, when=ta.crossunder(low, SL_line) or ta.crossover(high, TP_line) or status_none)
    strategy.exit('TP/SL','Base order', limit=TP_line, stop=SL_line, comment = Safe_order_line(100) > close ? 'SL' + i_s + ' - ' +  str.tostring(Base_order_size) + str.tostring(currency) : 'TP' + i_s + ' - ' +  str.tostring(Base_order_size) + str.tostring(currency)) 
    strategy.exit('TP/SL','Safety order' + i_s, limit=TP_line, stop=SL_line, comment = Safe_order_line(100) > close ? 'SL' + i_s + ' - ' +  str.tostring(safety_order_size(i)) + str.tostring(currency) : 'TP' + i_s + ' - ' +  str.tostring(safety_order_size(i)) + str.tostring(currency)) 
    // strategy.cancel('TP/SP' + i_s, when=Base_order_Condition)
    // strategy.exit('Stop Loss','Base order', stop=SL_line)
    // strategy.exit('Stop Loss','Safety order' + i_s, stop=SL_line)
    
//----------------label A----------------//

bot_usage(i) =>
 i == 1 ? Base_order_size + safety_order_size(1) :
 i == 2 ? Base_order_size + safety_order_size(1) + safety_order_size(2) :
 i == 3 ? Base_order_size + safety_order_size(1) + safety_order_size(2) + safety_order_size(3) :
 i == 4 ? Base_order_size + safety_order_size(1) + safety_order_size(2) + safety_order_size(3) + safety_order_size(4) : 
 i == 5 ? Base_order_size + safety_order_size(1) + safety_order_size(2) + safety_order_size(3) + safety_order_size(4) + safety_order_size(5) :
 i == 6 ? Base_order_size + safety_order_size(1) + safety_order_size(2) + safety_order_size(3) + safety_order_size(4) + safety_order_size(5) + safety_order_size(6) : 
 i == 7 ? Base_order_size + safety_order_size(1) + safety_order_size(2) + safety_order_size(3) + safety_order_size(4) + safety_order_size(5) + safety_order_size(6) + safety_order_size(7) : 
 i == 8 ? Base_order_size + safety_order_size(1) + safety_order_size(2) + safety_order_size(3) + safety_order_size(4) + safety_order_size(5) + safety_order_size(6) + safety_order_size(7) + safety_order_size(8) : 
 i == 9 ? Base_order_size + safety_order_size(1) + safety_order_size(2) + safety_order_size(3) + safety_order_size(4) + safety_order_size(5) + safety_order_size(6) + safety_order_size(7) + safety_order_size(8) + safety_order_size(9) :
 i == 10 ? Base_order_size + safety_order_size(1) + safety_order_size(2) + safety_order_size(3) + safety_order_size(4) + safety_order_size(5) + safety_order_size(6) + safety_order_size(7) + safety_order_size(8) + safety_order_size(9) + safety_order_size(10) : na

equity = strategy.equity
bot_use = bot_usage(Max_safety_trades_count)
bot_dev = float(step(Max_safety_trades_count)) * 100
bot_ava = (bot_use / equity) * 100

string label_A = 
 'Balance                                      : ' + str.tostring(math.round(equity, 0), '###,###,###,###') + ' USDT' + '\n' + 
 'Max amount for bot usage           : ' + str.tostring(math.round(bot_use, 0), '###,###,###,###') + ' USDT' + '\n' + 
 'Max safety order price deviation : ' + str.tostring(math.round(bot_dev, 0), '##.##') + ' %' + '\n' + 
 '% of available balance                : ' + str.tostring(math.round(bot_ava, 0), '###,###,###,###') + ' %' 
 + (bot_ava > 100 ? '\n \n' +  '⚠ Warning! Bot will use amount greater than you have on exchange' : na) 


if status_long
    day_label = 
     label.new(
     x=time[1], 
     y=high * 1.03, 
     text=label_A, 
     xloc=xloc.bar_time, 
     yloc=yloc.price, 
     color=bot_ava > 100 ? color.new(color.yellow, 0) : color.new(color.black, 50), 
     style=label.style_label_lower_right, 
     textcolor=bot_ava > 100 ? color.new(color.red, 0) : color.new(color.silver, 0), 
     size=size.normal, 
     textalign=text.align_left)

More