ChaoZhang
概述
该策略通过计算多个技术指标的综合信号,判断当前时间框架下的趋势方向。当判断为上涨趋势时,在较高点设定追踪止损线;当判断为下跌趋势时,在较低点设定追踪止损线。策略可以自适应不同品种和不同时间框架,通过动态调整止损线,实现风险控制。
原理
该策略结合了均线、ATR、KD、变动率等多个指标,判断当前时间框架下的总体趋势方向。具体来说,它计算出以下几个子信号的综合值:
- 均线方向信号
- KD指标超买超卖信号
- 量价背离信号
- 通道突破信号
- 多时间框架综合试错信号
- Percent R信号
- 均线回归信号
- ATR通道突破信号
上述每个子信号都经过了平滑处理,并设定不同的阈值判断买入/卖出。 然后对每个子信号进行加权,计算出当前时间框架下的总体信号。如果信号大于0,则判断为上涨趋势,如果信号小于0,则判断为下跌趋势。
在判断为上涨趋势时,策略会在之前较高点附近设定追踪止损线;在判断为下跌趋势时,策略会在之前较低点附近设定追踪止损线。这样可以根据实际价格走势来动态调整止损位,实现风险控制的目的。
优势
该策略集成了多个指标判断当前趋势方向,提高了判断的准确性。同时,策略可以自适应不同品种和时间框架,具有较强的适应性。
最重要的是,该策略能够动态调整止损线,能够根据实际走势调整风险控制水平,从而对冲系统性风险,这是其最大的优势。
风险
该策略判断趋势信号的质量直接影响止损线的设定,如果判断产生错误,可能导致止损位设定过于宽松或过于严格。此外,止损线无法完全规避行情突变的风险。
该策略还需要权衡获利水平和止损距离,如果止损距离过近,可能导致止损过于频繁;如果止损距离过远,则无法有效控制风险。这需要根据不同品种不同周期进行参数优化。
优化方向
可以考虑引入机器学习算法,利用历史数据训练判断趋势方向的模型,从而提高判断准确性。
可以测试不同参数组合,优化止损线的距离。例如动态调整ATR周期参数,以适应市场波动率的变化。
还可以结合交易量能量指标判断真实趋势,防止量价背离导致的信号错误。
总结
该策略通过集成多个技术指标判断当前趋势方向,并据此动态调整追踪止损线,旨在提高止损的实效性,控制交易风险。该策略理念先进,值得进一步优化和验证,是一个可供参考的多时间框架自适应风险控制策略。
策略源码
/*backtest
start: 2022-11-14 00:00:00
end: 2023-11-20 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/
// © jigneshjc
//@version=5
strategy("Jigga - Survival Level", shorttitle='Jigga - Survival Level', overlay=true)
doBackTesting = input(true, 'Run Back Testing')
entryCondition = false
exitCondition = false
ab21 = 14, gh41 = ab21
gh42 = ab21, ju51 = 14
ki61 = ju51
lkolkp = true ,ab22 = 58
cd31 = 5 , ab23 = 42
aa12 = 29, cd32 = 26
op71 = 5, aa11 = 12
aa13 = 9, op72 = 2.0
movnwx = false
kahachale(byju, h, l) =>
mika = ta.change(h)
awer = -ta.change(l)
uikmhDM = na(mika) ? na : mika > awer and mika > 0 ? mika : 0
wrtdfcDM = na(awer) ? na : awer > mika and awer > 0 ? awer : 0
bbct = ta.rma(ta.tr, byju)
uikmh = fixnan(100 * ta.rma(uikmhDM, byju) / bbct)
wrtdfc = fixnan(100 * ta.rma(wrtdfcDM, byju) / bbct)
[uikmh, wrtdfc]
trial(gh42, gh41, h, l) =>
[uikmh, wrtdfc] = kahachale(gh42, h, l)
uuolop = uikmh + wrtdfc
trial = 100 * ta.rma(math.abs(uikmh - wrtdfc) / (uuolop == 0 ? 1 : uuolop), gh41)
trial
_pr(src, byjugth) =>
max = ta.highest(byjugth)
min = ta.lowest(byjugth)
100 * (src - max) / (max - min)
kyukarna(khulmkhula, mikaarwala, nichewala, bandhwala, partiwala) =>
sig = trial(gh42, gh41, mikaarwala, nichewala)
trialIncreasing = sig > ta.ema(sig, 5) ? lkolkp : movnwx
rolkmn = ta.ema(bandhwala, aa11)
psolkmn = ta.ema(bandhwala, aa12)
ujghd = rolkmn - psolkmn
wrtycv = ta.ema(ujghd, aa13)
kimnjg = ujghd - wrtycv
mikalilo = ta.rma(math.max(ta.change(bandhwala), 0), ab21)
awerlilo = ta.rma(-math.min(ta.change(bandhwala), 0), ab21)
lilo = awerlilo == 0 ? 100 : mikalilo == 0 ? 0 : 100 - 100 / (1 + mikalilo / awerlilo)
juylknlilo = ta.ema(lilo, 3)
rjuylkn = ta.ema(bandhwala, cd31)
psjuylkn = ta.ema(bandhwala, cd32)
percentR = _pr(bandhwala, ju51)
juylknpercentR = ta.ema(percentR, 3)
ad = bandhwala == mikaarwala and bandhwala == nichewala or mikaarwala == nichewala ? 0 : (2 * bandhwala - nichewala - mikaarwala) / (mikaarwala - nichewala) * partiwala
kiloValue = math.sum(ad, ki61) / math.sum(partiwala, ki61)
liiopn = ta.atr(op71)
mikaliiopn = (mikaarwala + nichewala) / 2 - op72 * liiopn
mika1liiopn = nz(mikaliiopn[1], mikaliiopn)
mikaliiopn := bandhwala[1] > mika1liiopn ? math.max(mikaliiopn, mika1liiopn) : mikaliiopn
dnliiopn = (mikaarwala + nichewala) / 2 + op72 * liiopn
dn1liiopn = nz(dnliiopn[1], dnliiopn)
dnliiopn := bandhwala[1] < dn1liiopn ? math.min(dnliiopn, dn1liiopn) : dnliiopn
omnerliiopn = 1
omnerliiopn := nz(omnerliiopn[1], omnerliiopn)
omnerliiopn := omnerliiopn == -1 and bandhwala > dn1liiopn ? 1 : omnerliiopn == 1 and bandhwala < mika1liiopn ? -1 : omnerliiopn
fitur = ujghd > 0 ? ujghd > wrtycv ? 1 : 0 : ujghd > wrtycv ? 0 : -1
mitur = kimnjg >= 0 ? kimnjg > kimnjg[1] ? 1 : 0 : kimnjg > kimnjg[1] ? 0 : -1
ritur = juylknlilo > ab22 ? 1 : juylknlilo < ab23 ? -1 : 0
circuits = rjuylkn > psjuylkn ? 1 : -1
trialPoints = trialIncreasing ? close > ta.ema(close, 3) ? 1 : -1 : 0
virar = juylknpercentR > -ab23 ? 1 : juylknpercentR < -ab22 ? -1 : 0
chikar = kiloValue > 0.1 ? 1 : kiloValue < -0.1 ? -1 : 0
sitar = omnerliiopn
p = fitur + mitur + ritur + circuits + trialPoints + virar + chikar + sitar
p
currentP = kyukarna(open, high, low, close, volume)
currentPNew = currentP >= 0 and currentP[1] <= 0 ? 0 : currentP <= 0 and currentP[1] >= 0 ? 0 : currentP
colorPNew = currentPNew == 0 ? color.black : currentPNew >= 0 ? color.green : color.red
//plot(currentPNew, color=colorPNew, title='CurrentTimeFrame')
LTN = 0.0
LTN := nz(LTN) ? 0.0 : (currentPNew[1] < 0 and currentPNew >= 0) ? high * 1.005 : (currentPNew[1] > 0 and currentPNew <= 0) ? low * 0.995 : LTN[1]
LClr = color.green
LClr := (currentPNew[1] < 0 and currentPNew >= 0) ? color.green : (currentPNew[1] > 0 and currentPNew <= 0) ? color.red : LClr[1]
plot(LTN,color=LClr,title="Level", style=plot.style_circles)
entryCondition:= high > LTN and LClr == color.green ? lkolkp : movnwx
exitCondition:= low < LTN and LClr == color.red ? lkolkp : movnwx
tradeRunning = movnwx
tradeRunning := nz(tradeRunning) ? movnwx : (not tradeRunning[1]) and entryCondition ? lkolkp : tradeRunning[1] and exitCondition ? movnwx : tradeRunning[1]
plotshape(tradeRunning and (not tradeRunning[1]) and (not doBackTesting), style=shape.labelup, location=location.belowbar, color=color.new(#00FF00, 50), size=size.tiny, title='Buy wrtycv', text='➹', textcolor=color.new(color.black,0))
plotshape((not tradeRunning) and tradeRunning[1] and (not doBackTesting), style=shape.labeldown, location=location.abovebar, color=color.new(#FF0000, 50), size=size.tiny, title='Sell wrtycv', text='➷', textcolor=color.new(color.white, 0))
if entryCondition and doBackTesting
strategy.entry(id="Buy",direction=strategy.long)
if exitCondition and doBackTesting
strategy.close(id="Buy")