资源加载中... loading...

Dynamic Channel Breakthrough Strategy

Author: ChaoZhang, Date: 2024-02-27 15:15:07
Tags:

img

Overview

This strategy uses the Keltner Channel indicator, combined with moving average lines, to set dynamic breakout buy and sell prices to achieve low-buy-high-sell breakthrough operations. The strategy can automatically identify channel breakout buy and sell opportunities.

Strategy Principle

  1. Calculate channel median: Use exponential moving average to calculate the price median of the channel
  2. Calculate channel bandwidth: Use the moving average of true volatility, average true volatility or price amplitude to calculate channel bandwidth
  3. Channel upper and lower rail: Median ± N times channel bandwidth
  4. Entry order: When the price touches the upper rail, set the breakthrough buy price and wait for the breakthrough; when the price touches the lower rail, set the breakthrough sell price and wait for the breakthrough
  5. Exit order: Stop loss when price falls back to median after buying, or when highest price exceeds entry price; Stop loss when price bounces back to median after selling, or when lowest price is lower than entry price

Advantage Analysis

  1. Using dynamic channels can quickly capture changes in market trends
  2. Using median is conducive to judging the direction of price trends
  3. N times bandwidth setting makes the channel range reasonable to avoid frequent position adjustments
  4. Using breakthrough mechanisms conforms to trend theory and follows the trend
  5. Setting stop loss conditions strictly controls risks

Risk Analysis

  1. The selection of the method for calculating the median line will affect the matching effect of the channel range and prices
  2. Excessive large or small N multiples will affect the strategy’s rate of return
  3. Breakthrough buys and sells tend to form false signals and should be strictly stopped out

Optimization Directions

  1. Try different median line calculation methods to find the optimal parameters
  2. Test different N values to find the optimal multiplier
  3. Increase the breakthrough amplitude to avoid false signals
  4. Optimize stop loss logic to strictly control single loss

Summary

The overall strategy uses scientific and reasonable methods to judge price trends and directions through dynamic channel indicators, sets reasonable parameters to capture breakthrough signals, achieves low-buy-high-sell, and gains excess returns. At the same time, continuously optimize the risks of the strategy so that it can run stably in various markets.


/*backtest
start: 2024-01-27 00:00:00
end: 2024-02-26 00:00:00
period: 1h
basePeriod: 15m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/

//@version=5
strategy(title="Keltner Strategy", overlay=true)
length = input.int(20, minval=1)
mult = input.float(2.0, "Multiplier")
src = input(close, title="Source")
exp = input(true, "Use Exponential MA")
BandsStyle = input.string("Average True Range", options = ["Average True Range", "True Range", "Range"], title="Bands Style")
atrlength = input(10, "ATR Length")
esma(source, length)=>
	s = ta.sma(source, length)
	e = ta.ema(source, length)
	exp ? e : s
ma = esma(src, length)
rangema = BandsStyle == "True Range" ? ta.tr(true) : BandsStyle == "Average True Range" ? ta.atr(atrlength) : ta.rma(high - low, length)
upper = ma + rangema * mult
lower = ma - rangema * mult
crossUpper = ta.crossover(src, upper)
crossLower = ta.crossunder(src, lower)
bprice = 0.0
bprice := crossUpper ? high+syminfo.mintick : nz(bprice[1])
sprice = 0.0
sprice := crossLower ? low -syminfo.mintick : nz(sprice[1])
crossBcond = false
crossBcond := crossUpper ? true
     : na(crossBcond[1]) ? false : crossBcond[1]
crossScond = false
crossScond := crossLower ? true
     : na(crossScond[1]) ? false : crossScond[1]
cancelBcond = crossBcond and (src < ma or high >= bprice )
cancelScond = crossScond and (src > ma or low <= sprice )
if (cancelBcond)
	strategy.cancel("KltChLE")
if (crossUpper)
	strategy.entry("KltChLE", strategy.long, stop=bprice, comment="KltChLE")
if (cancelScond)
	strategy.cancel("KltChSE")
if (crossLower)
	strategy.entry("KltChSE", strategy.short, stop=sprice, comment="KltChSE")

More