Low Pyramid Risk Tracking Strategy

Author: ChaoZhang, Date: 2023-12-22 12:56:36
Tags:

img

This strategy identifies potential low points in price movement through a combination of different indicators and gradually builds positions through pyramiding to reduce risk. The strategy also incorporates functions such as stop loss, take profit, and trailing stop loss to effectively control risk.

Strategy Overview

The strategy first uses the difference between RSI and EMA RSI to identify potential price lows. To filter out false signals, the strategy also combines moving average and multi-timeframe stochastic indicator for confirmation. Once the low point signal is confirmed, long positions will be gradually built at slightly lower prices from that point through pyramiding. The strategy allows up to 12 tracking orders to be opened, with the size of each order increasing in sequence, which can effectively diversify risks. All orders will follow an overall stop loss to exit, while allowing to set take profit separately for each order. To further control risks, the strategy also sets an overall stop loss based on equity percentage.

Strategy Principle

The strategy consists of three main modules: low point identification, pyramid tracking and risk control.

The low point identification module uses the difference between RSI and its EMA to identify potential price lows. To improve accuracy, moving average indicator and multi-timeframe stochastic indicators are introduced for signal filtering. Only when price is below moving average and stochastic K line is below 30, the validity of the low point signal will be confirmed.

The pyramid tracking module is the core of this strategy. Once the low point signal is confirmed, the strategy will open the first position at 0.1% below that low point. Afterwards, as long as price keeps falling and is below certain percentage of average entry price, more long orders will be added. The size of new orders will increase in sequence, for example the third order is 3 times the first order size. This pyramid tracking approach helps averaging risks. The strategy allows up to 12 tracking orders.

The risk control module includes three aspects. First is the overall stop loss based on highest price in recent periods. All orders will follow this stop loss. Second is independent take profit setting for each order, which allows to close order based on certain percentage of entry price. Third is overall stop loss based on percentage of account equity, which is the strongest risk control method.

Strategy Advantages

  • Pyramid tracking reduces risk of individual orders while diversifying overall risk
  • Combination of indicators improves accuracy of low point identification
  • Overall stop loss, take profit and trailing stop functions effectively control risk
  • Equity stop loss protects account from significant losses
  • Parameters can be tuned to balance risk vs reward

Strategy Risks

  • Low point identification still has some limitation, may miss best entry point or get into false signal
  • Facing adverse market when adding orders may increase loss
  • Needs relatively long period to reflect the advantage
  • Inappropriate parameter setting may lead to insufficient risk control

To reduce above risks, some aspects can be optimized:

  1. Change or add indicators to improve low point identification accuracy
  2. Optimize number of orders, intervals, take profit percentage etc to lower risk per order
  3. Moderately tighten stop loss level to protect profits
  4. Test different products with good liquidity and large fluctuation

Strategy Optimization

There is still room for further optimization of this strategy:

  1. Try introducing more advanced techniques like machine learning for low point identification
  2. Dynamically adjust order quantity, stop loss level etc based on market condition
  3. Add box stop loss mechanism to avoid expanding losses
  4. Add re-entry mechanism
  5. Optimize parameters for stocks and crypto currencies

Summary

This strategy effectively reduces risks of individual orders through pyramid tracking approach, and overall stop loss, take profit, trailing stop functions also play very good role of risk control. But there is still room for improving low point identification and other aspects. If more advanced techniques can be introduced, dynamic adjustment of parameters can be added, combined with parameter optimization, the risk reward ratio of this strategy could be greatly improved.


/*backtest
start: 2022-12-15 00:00:00
end: 2023-12-21 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/
// © A3Sh
//@version=5

// Strategy that finds potential lows in the price action and spreads the risk by entering multiple positions at these lows.
// The low is detected based on the difference between MTF RSI and EMA based RSI, Moving Average and MTF Stochastic indicators.
// The size of each next position that is entered is multiplied by the sequential number of the position.
// Each separate position can exit when a specified take profit is triggered and re-open when detecting a new potential low.
// All positions are closed when the price action crosses over the dynamic blue stop level line.

// This strategy combines open-source code developed by fellow Tradingview community members: 
// The Lowfinder code is developed by RafaelZioni
// https://www.tradingview.com/script/GzKq2RVl-Low-finder/
// Both the MTF RSI code and the MTF Stochastic code are adapted from the MTFindicators libary written by Peter_O
// https://www.tradingview.com/script/UUVWSpXR-MTFindicators/

// The Stop Level calculation is inspired by the syminfo-mintick tutorial on Kodify.net
// https://kodify.net/tradingview/info/syminfo-mintick/

strategy("LowFinder_PyraMider", 
         overlay=true, pyramiding=99, 
         precision=2,
         initial_capital=10000, 
         default_qty_type=strategy.percent_of_equity, 
         default_qty_value=10,
         commission_type=strategy.commission.percent, 
         commission_value=0.06,
         slippage=1
         )


// Backtest Window
start_time   = input(defval=timestamp("01 April 2021 20:00"), group = "Backtest Window", title="Start Time")
end_time     = input(defval=timestamp("01 Aug 2030 20:00"),   group = "Backtest Window", title="End Time")
window() => true


// Inputs
portfolio_size  = input.float  (100,         group = 'Risk - Portfolio',       title = 'Portfolio %', step=1.0) / 100
leverage        = input.int    (1,           group = 'Risk - Portfolio',       title = 'Leverage', minval = 1)
q_mode          = input.string ('multiply',  group = 'Risk - Order Size',      title = 'Order Size Mode', options = ['base', 'multiply'], tooltip = 'Base mode: the base quantiy for each sequential order. Multiply mode: each quantity is multiplied by order number')
q_mode_m        = input.int    (1,           group = 'Risk - Order Size',      title = 'Order Size Divider (Multiply Mode)',  tooltip = 'Divide Multiply by this number to lower the sequential order sizes')
fixed_q         = input.bool   (false,       group = 'Risk - Order Size',      title = 'Fixed Order Size',     inline = '01', tooltip = 'Use with caution! Overrides all Risk calculations')
amount_q        = input.float  (1,           group = 'Risk - Order Size',      title = '. . Base Currency:',   inline = '01')
sl_on           = input.bool   (false,       group = 'Risk - Stop Loss',       title = 'StopLoss of',          inline = '03')
stopLoss        = input.float  (1.5,         group = 'Risk - Stop Loss',       title = '',   step=0.1,         inline = '03') / 100
sl_mode         = input.string ('equity',    group = 'Risk - Stop Loss',       title = '% of',  options = ['avg_price', 'equity'], inline = '03')
stop_len        = input.int    (100,         group = 'Risk - Stop Level',      title = 'Stop Level Length', tooltip = 'Lookback most recent highest high')
stop_deviation  = input.float  (0.3,         group = 'Risk - Stop Level',      title = 'Deviatation % above Stop Level', step=0.1) / 100
cond2_toggle    = input.bool   (true ,       group = 'Risk - Take Profit',     title = 'Take Profit/Trailing Stop', inline = '04')
tp_all          = input.float  (1.0,         group = 'Risk - Take Profit',     title = '..........%', step=0.1,     inline = '04') / 100
tp_on           = input.bool   (true,        group = 'Risk - Take Profit',     title = 'Exit Crossover Take Profit and .....', inline = '02')
exit_mode       = input.string ('stoplevel', group = 'Risk - Take Profit',     title = '',   options = ['close', 'stoplevel'], inline = '02')
takeProfit      = input.float  (10.0,        group = 'Risk - Take Profit',     title = 'Take Profit % per Order', tooltip = 'Each separate order exits when hit', step=0.1)
posCount        = input.int    (12,          group = 'Pyramiding Settings',    title = 'Max Number of Orders')
next_entry      = input.float  (0.2,         group = 'Pyramiding Settings',    title = 'Next Order % below Avg. Price', step=0.1)
oa_lookback     = input.int    (0,           group = 'Pyramiding Settings',    title = 'Next Order after X candles', tooltip = 'Prevents opening too much orders in a Row')
len_rsi         = input.int    (5,           group = 'MTF LowFinder Settings', title = 'Lookback of RSI')
mtf_rsi         = input.int    (1,           group = 'MTF LowFinder Settings', title = 'Higher TimeFrame Multiplier RSI',  tooltip='Multiplies the current timeframe by specified value')
ma_length       = input.int    (26,          group = 'MTF LowFinder Settings', title = 'MA Length / Sensitivity')
new_entry       = input.float  (0.1,         group = 'MTF LowFinder Settings', title = 'First Order % below Low',step=0.1, tooltip = 'Open % lower then the found low')/100
ma_signal       = input.int    (100,         group = 'Moving Average Filter',  title = 'Moving Average Length')
periodK         = input.int    (14,          group = 'MTF Stochastic Filter',  title = 'K',      minval=1)
periodD         = input.int    (3,           group = 'MTF Stochastic Filter',  title = 'D',      minval=1)
smoothK         = input.int    (3,           group = 'MTF Stochastic Filter',  title = 'Smooth', minval=1)
lower           = input.int    (30,          group = 'MTF Stochastic Filter',  title = 'MTF Stoch Filter (above gets filtered)')
mtf_stoch       = input.int    (10,          group = 'MTF Stochastic Filter',  title = 'Higher TimeFrame Multiplier', tooltip='Multiplies the current timeframe by specified value')
avg_on          = input.bool   (true,        group = 'Plots',                  title = 'Plot Average Price')
plot_ma         = input.bool   (false,       group = 'Plots',                  title = 'Plot Moving Average')
plot_ts         = input.bool   (false,       group = 'Plots',                  title = 'Plot Trailing Stop Level')


// variables //
var entry_price     = 0.0    // The entry price of the first entry
var previous_entry  = 0.0    // Stores the price of the previous entry
var iq              = 0.0    // Inititial order quantity before risk calculation
var nq              = 0.0    // Updated new quantity after the loop
var oq              = 0.0    // Old quantity at the beginning or the loop
var q               = 0.0    // Final calculated quantity used as base order size
var int order_after = 0



// Order size calaculations // 

// Order size based on max amount of pyramiding orders or fixed by user input ///
// Order size calculation based on 'base' mode or ' multiply' mode //
if fixed_q
    q := amount_q
else if q_mode == 'multiply'
    iq := (math.abs(strategy.equity * portfolio_size  / posCount) / open) * leverage
    oq := iq
    for i = 0 to posCount
        nq := oq + (iq * ( i/ q_mode_m + 1))
        oq := nq 
    q := (iq  * posCount /  oq) * iq

else
    q := (math.abs(strategy.equity * portfolio_size  / posCount) / open) * leverage

// Function to calcaulate final order size based on order size modes and round the result with 1 decimal //
quantity_mode(index,string q_mode) =>
    q_mode == 'base' ? math.round(q,1) : q_mode == 'multiply' ? math.round(q * (index/q_mode_m  + 1),1) : na



// LowFinder Calculations //
// MTF RSI by Peter_O //
rsi_mtf(float source, simple int mtf,simple int len) =>
    change_mtf=source-source[mtf]
    up_mtf = ta.rma(math.max(change_mtf, 0), len*mtf)
    down_mtf = ta.rma(-math.min(change_mtf, 0), len*mtf)
    rsi_mtf = down_mtf == 0 ? 100 : up_mtf == 0 ? 0 : 100 - (100 / (1 + up_mtf / down_mtf))

// Lowfinder by RafaelZioni //
vrsi = rsi_mtf(close,mtf_rsi,len_rsi)

pp=ta.ema(vrsi,ma_length)
dd=(vrsi-pp)*5
cc=(vrsi+dd+pp)/2

lows=ta.crossover(cc,0) 



// MTF Stoch Calcualation // MTF Stoch adapted from  Peter_O //
stoch_mtfK(source, mtf, len) =>

    k = ta.sma(ta.stoch(source, high, low, periodK * mtf), smoothK * mtf)
    
stoch_mtfD(source, mtf, len) =>

    k = ta.sma(ta.stoch(source, high, low, periodK * mtf), smoothK * mtf)
    d = ta.sma(k, periodD * mtf)
    
mtfK = stoch_mtfK(close, mtf_stoch, periodK)
mtfD = stoch_mtfD(close, mtf_stoch, periodK)



// Open next position % below average position price //
below_avg = close < (strategy.position_avg_price * (1 - (next_entry / 100)))



// Moving Average Filter //
moving_average_signal = ta.sma(close, ma_signal)
plot (plot_ma ? moving_average_signal : na, title = 'Moving Average', color = color.rgb(154, 255, 72))



// Buy Signal //
buy_signal = lows and close < moving_average_signal and mtfK < lower
// First Entry % Below lows //
if buy_signal
    entry_price := close * (1 - new_entry)



// Plot Average Price of Position//
plot (avg_on  ? strategy.position_avg_price : na, title = 'Average Price', style = plot.style_linebr, color = color.new(color.white,0), linewidth = 1)



// Take profit per Open Order //
take_profit_price = close * takeProfit / 100 / syminfo.mintick



// Calculate different Stop Level conditions to exit All //

// Stop Level Caculation //
stop_long1_level = ta.highest (high, stop_len)[1]  * (1 + stop_deviation)
stop_long2_level = ta.highest (high, stop_len)[2]  * (1 + stop_deviation)
stop_long3_level = ta.highest (high, stop_len)[3]  * (1 + stop_deviation)
stop_long4_level = ta.highest (high, stop_len)[1]  * (1 - 0.008) 
// Stop triggers //
stop_long1 = ta.crossover(close,stop_long1_level)
stop_long2 = ta.crossover(close,stop_long2_level)
stop_long4 = ta.crossunder(close,stop_long4_level)
// Exit Conditions, cond 1 only Stop Level, cond2 Trailing Stop option //
exit_condition_1 = close < strategy.position_avg_price ? stop_long1 : close > strategy.position_avg_price ? stop_long2 : na
exit_condition_2 = close < strategy.position_avg_price * (1 + tp_all) ? stop_long2 : 
                   close > strategy.position_avg_price * (1 + tp_all) ? stop_long4 :
                   close < strategy.position_avg_price ? stop_long1 : na
// Switch between conditions //
exit_conditions = cond2_toggle ? exit_condition_2 : exit_condition_1

// Exit when take profit //
ex_m = exit_mode == 'close' ? close : stop_long2_level
tp_exit = ta.crossover(ex_m, strategy.position_avg_price * (1 + tp_all)) and close > strategy.position_avg_price * 1.002

// Plot stoplevel, take profit level //
plot_stop_level    = strategy.position_size > 0 ? stop_long2_level : na
plot_trailing_stop = cond2_toggle and plot_ts and strategy.position_size > 0 and close > strategy.position_avg_price * (1 + tp_all) ? stop_long4_level : na

plot(plot_stop_level,    title = 'Stop Level',    style=plot.style_linebr, color = color.new(#41e3ff, 0), linewidth = 1)
plot(plot_trailing_stop, title = 'Trailing Stop', style=plot.style_linebr, color = color.new(#4cfca4, 0), linewidth = 1)

plot_tp_level = cond2_toggle and strategy.position_size > 0 ? strategy.position_avg_price * (1 + tp_all) : na
plot(plot_tp_level, title = 'Take Profit Level', style=plot.style_linebr, color = color.new(#ff41df, 0), linewidth = 1)



// Calculate Stop Loss based on equity and average price //
loss_equity = ((strategy.position_size * strategy.position_avg_price) - (strategy.equity * stopLoss)) / strategy.position_size
loss_avg_price = strategy.position_avg_price * (1 - stopLoss)
stop_loss = sl_mode == 'avg_price' ? loss_avg_price : loss_equity
plot(strategy.position_size > 0  and sl_on ? stop_loss : na, title = 'Stop Loss', color=color.new(color.red,0),style=plot.style_linebr, linewidth = 1)



// Enter first position //
if ta.crossunder(close,entry_price) and window() and strategy.position_size == 0
    strategy.entry('L_1', strategy.long, qty = math.round(q,1), comment = '+' + str.tostring(math.round(q,1)))
    previous_entry := close


// Enter next pyramiding positions //
if buy_signal and window() and strategy.position_size > 0 and below_avg
    order_after := order_after + 1
    for i = 1 to strategy.opentrades
        entry_comment = '+' + str.tostring((quantity_mode(i,q_mode))) // Comment with variable //
        if strategy.opentrades == i and i < posCount and order_after > oa_lookback
            entry_price := close
            entry_id = 'L_' + str.tostring(i + 1) 
            strategy.entry(id = entry_id, direction=strategy.long, limit=entry_price, qty= quantity_mode(i,q_mode), comment = entry_comment)
            previous_entry := entry_price
            order_after := 0


// Exit per Position //
if strategy.opentrades > 0 and window() 
    for i = 0 to strategy.opentrades 
        exit_comment = '-' + str.tostring(strategy.opentrades.size(i))
        exit_from = 'L_' + str.tostring(i + 1)
        exit_id = 'Exit_' + str.tostring(i + 1)
        strategy.exit(id= exit_id, from_entry= exit_from, profit = take_profit_price, comment = exit_comment)
            

// Exit All //
if exit_conditions or (tp_exit and tp_on and cond2_toggle) and window()
    strategy.close_all('Exti All')
    entry_price := 0

if ta.crossunder(close,stop_loss)  and sl_on and window()
    strategy.close_all('StopLoss')
    entry_price := 0
    





More