Estratégia de negociação de impulso médio móvel de vários prazos

Autora:ChaoZhang, Data: 2023-11-06 16:04:46
Tags:

Resumo

Esta estratégia de negociação combina várias médias móveis e indicadores de impulso para identificar a direção e a força das tendências, estabelecendo posições no início das tendências emergentes e, em seguida, otimizando o gerenciamento de lucros e riscos através do uso de trailing stops e take profits.

Estratégia lógica

Dois conjuntos de médias móveis com parâmetros diferentes são construídos para criar linhas rápidas e lentas:
- A linha rápida consiste numa EMA de 5 períodos e numa WMA de 25 períodos, representando uma tendência de curto prazo.
- A linha lenta consiste numa EMA de 28 períodos e numa WMA de 72 períodos, representando uma tendência de médio a longo prazo
Quando a linha rápida cruza acima da linha lenta, sinaliza que a tendência de curto prazo está ganhando força sobre a tendência de médio prazo e apresenta sinal de entrada.
Filtro RSI incorporado, tomando apenas entradas em mínimos do RSI (para longo) e máximos do RSI (para curto) para evitar falhas.
Uma vez inserido, o trailing stop é usado para minimizar as perdas, enquanto o take profit bloqueia os ganhos.
Quando a linha rápida cruza abaixo da linha lenta, ela sinaliza a reversão da tendência, solicitando a saída via stop loss ou take profit.

Análise das vantagens

As combinações de médias móveis duplas filtram o ruído e identificam a direção e a força da tendência durante a progressão da tendência.
As entradas estabelecidas precocemente nas tendências emergentes evitam perdas desnecessárias por causa de falhas.
O filtro RSI melhora a qualidade das entradas.
O trailing stop comprime as perdas de negociações perdedoras individuais.
Tome lucro permite a captura de lucro considerável durante movimentos favoráveis.

Análise de riscos

As médias móveis duplas podem atrasar-se em pontos de virada, perdendo oportunidades de reversão.
- Redução dos períodos da média móvel para maior sensibilidade.
Falsas fugas levam a entradas desnecessárias.
- Incorpore mais filtros antes de entrar.
As distâncias stop loss/take profit não são otimizadas, podem ser demasiado largas ou demasiado estreitas.
- Otimizar através de backtest para encontrar níveis ideais.
A estratégia direcional só funciona para mercados em tendência.
- Apenas empregue estratégias baseadas nas condições do mercado.

Orientações de otimização

Otimizar os parâmetros da média móvel para encontrar as melhores representações da tendência.
Adicionar filtros de tendência como parada ATR dinâmica, osciladores de momento, etc.
Otimizar os parâmetros de stop loss / take profit.
Incorporar filtros de regime baseados nas condições de mercado.
Adicionar análise de quadros de tempo cruzados usando quadros de tempo mais longos para orientar a direção da estratégia de curto prazo.

Resumo

Esta estratégia integra médias móveis e indicadores de momento para identificar e estabelecer entradas iniciais durante tendências emergentes, enquanto gerencia risco e recompensa através de paradas oportunas e ganhos. Embora seja necessária uma otimização adicional de parâmetros e lógica para se adaptar às condições mais amplas do mercado, ela já possui uma estrutura básica e direcionalidade para capturar tendências de médio e longo prazo. Com melhorias contínuas, esta estratégia pode potencialmente se tornar um sistema robusto e eficiente de acompanhamento de tendências.

/*backtest
start: 2023-10-06 00:00:00
end: 2023-11-05 00:00:00
period: 1h
basePeriod: 15m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/

//@version=3
strategy(title="[WAI GUA]", shorttitle="[EOS] 1.0", overlay=true)

//study(title="[WAI GUA]", shorttitle="[EOS] 1.0", overlay=true)



//
// Use Alternate Anchor TF for MAs 
uRenko    = input(true, title="IS This a RENKO Chart")
//
anchor     = input(0,minval=0,maxval=1440,title="Alternate TimeFrame Multiplier (0=none)")
//
src          = close //input(close, title="EMA Source")
showRibbons  = input(false,title="Show Coloured MA Ribbons")
showAvgs     = input(false,title="Show Ribbon Median MA Lines")

//
// Fast Ribbon MAs
// Lower MA - type, length
typeF1    = input(defval="EMA", title="FAST MA Ribbon Type: ", options=["SMA", "EMA", "WMA", "VWMA", "SMMA", "DEMA", "TEMA", "LAGMA", "HullMA", "ZEMA", "TMA", "SSMA"])
lenF1     = input(defval=5, title="FAST Ribbon Lower MA Length", minval=1)
gammaF1   = 0.33 //input(defval=0.33,title="Fast MA - Gamma for LAGMA")

// Upper MA - type, length
typeF11   = typeF1 //input(defval="WMA", title="FAST Ribbon Upper MA Type: ", options=["SMA", "EMA", "WMA", "VWMA", "SMMA", "DEMA", "TEMA", "LAGMA", "HullMA", "ZEMA", "TMA", "SSMA"])
lenF11    = input(defval=25, title="FAST Ribbon Upper Length", minval=2)
gammaF11  = 0.77 //input(defval=0.77,title="Slow MA - Gamma for LAGMA")

// Slow Ribbon MAs
// Lower MA - type, length
typeS1   = input(defval="EMA", title="SLOW MA Ribbon Type: ", options=["SMA", "EMA", "WMA", "VWMA", "SMMA", "DEMA", "TEMA", "LAGMA", "HullMA", "ZEMA", "TMA", "SSMA"])
lenS1    = input(defval=28, title="SLOW Ribbon Lower MA Length", minval=1)
gammaS1  = 0.33 //input(defval=0.33,title="Fast MA - Gamma for LAGMA")

// Upper MA - type, length
typeS16   = typeS1 //input(defval="WMA", title="SLOW Ribbon Upper MA Type: ", options=["SMA", "EMA", "WMA", "VWMA", "SMMA", "DEMA", "TEMA", "LAGMA", "HullMA", "ZEMA", "TMA", "SSMA"])
lenS16    = input(defval=72, title="SLOW Ribbon Upper Length", minval=2)
gammaS16  = 0.77 //input(defval=0.77,title="Slow MA - Gamma for LAGMA")

// - Constants
gold = #FFD700

// - FUNCTIONS
// - variant(type, src, len, gamma)
// 返回MA输入选择变量，默认为SMA，如果空白或键入。

// SuperSmoother filter
variant_supersmoother(src,len) =>
    a1 = exp(-1.414*3.14159 / len)
    b1 = 2*a1*cos(1.414*3.14159 / len)
    c2 = b1
    c3 = (-a1)*a1
    c1 = 1 - c2 - c3
    v9 = 0.0
    v9 := c1*(src + nz(src[1])) / 2 + c2*nz(v9[1]) + c3*nz(v9[2])
    v9
    
variant_smoothed(src,len) =>
    v5 = 0.0
    v5 := na(v5[1]) ? sma(src, len) : (v5[1] * (len - 1) + src) / len
    v5

variant_zerolagema(src,len) =>
    ema1 = ema(src, len)
    ema2 = ema(ema1, len)
    v10 = ema1+(ema1-ema2)
    v10
    
variant_doubleema(src,len) =>
    v2 = ema(src, len)
    v6 = 2 * v2 - ema(v2, len)
    v6

variant_tripleema(src,len) =>
    v2 = ema(src, len)
    v7 = 3 * (v2 - ema(v2, len)) + ema(ema(v2, len), len)               // Triple Exponential
    v7
    
//calc Laguerre
variant_lag(p,g) =>
    L0 = 0.0
    L1 = 0.0
    L2 = 0.0
    L3 = 0.0
    L0 := (1 - g)*p+g*nz(L0[1])
    L1 := -g*L0+nz(L0[1])+g*nz(L1[1])
    L2 := -g*L1+nz(L1[1])+g*nz(L2[1])
    L3 := -g*L2+nz(L2[1])+g*nz(L3[1])
    f = (L0 + 2*L1 + 2*L2 + L3)/6
    f

// return variant, defaults to SMA 
variant(type, src, len, g) =>
    type=="EMA"     ? ema(src,len) : 
      type=="WMA"   ? wma(src,len): 
      type=="VWMA"  ? vwma(src,len) : 
      type=="SMMA"  ? variant_smoothed(src,len) : 
      type=="DEMA"  ? variant_doubleema(src,len): 
      type=="TEMA"  ? variant_tripleema(src,len): 
      type=="LAGMA" ? variant_lag(src,g) :
      type=="HullMA"? wma(2 * wma(src, len / 2) - wma(src, len), round(sqrt(len))) :
      type=="SSMA"  ? variant_supersmoother(src,len) : 
      type=="ZEMA"  ? variant_zerolagema(src,len) : 
      type=="TMA"   ? sma(sma(src,len),len) : 
                      sma(src,len)

// - /variant 

// If have anchor specified, calculate the base multiplier.
//mult  = isintraday ? anchor==0 or interval<=0 or interval>=anchor or anchor>1440? 1 : round(anchor/interval) : 1
//mult := isdwm?  1 : mult  // Only available Daily or less
mult = anchor>0 ? anchor : 1 

//
high_  = uRenko? max(close,open) : high
low_   = uRenko? min(close,open) : low


//用锚乘器调整MA长度

//Fast MA Ribbon
emaF1 = variant(typeF1, src, lenF1*mult, gammaF1)
emaF11 = variant(typeF11, src, lenF11*mult,gammaF11)
emafast = (emaF1+emaF11)/2 // Average of Upper and Lower MAs
//
//Slow MA Ribbon
emaS1 = variant(typeS1,src, lenS1*mult,gammaS1)
emaS16 = variant(typeS16, src, lenS16*mult, gammaS16)
emaslow = (emaS1+emaS16)/2 // Average of Upper and Lower MAs
//
// Count crossover candles
xup = 0
xdn = 0
fup = 0
fdn = 0
sup = 0
sdn = 0
// 
xup := (emafast-emaslow)>0 and (emafast-emaslow)>(emafast[1]-emaslow[1]) ? nz(xup[1])+1 : 0
xdn := (emafast-emaslow)<0 and (emafast-emaslow)<(emafast[1]-emaslow[1]) ? nz(xdn[1])+1 : 0
fup := (emaF1-emaF11)>0 and (emaF1-emaF11)>(emaF1[1]-emaF11[1]) ? nz(fup[1])+1 : 0
fdn := (emaF1-emaF11)<0 and (emaF1-emaF11)<(emaF1[1]-emaF11[1]) ? nz(fdn[1])+1 : 0
sup := (emaS1-emaS16)>0 and (emaS1-emaS16)>(emaS1[1]-emaS16[1]) ? nz(sup[1])+1 : 0
sdn := (emaS1-emaS16)<0 and (emaS1-emaS16)<(emaS1[1]-emaS16[1]) ? nz(sdn[1])+1 : 0

//Fast EMA Final Color Rules
colFinal = fup>=2 ? aqua : fdn>=2 ? blue : gray
//Slow EMA Final Color Rules
colFinal2 = sup>=2 ? lime : sdn>=2 ? red : gray

//Fast EMA Plots
p1=plot(showRibbons?emaF1:na, title="Fast Ribbon Lower MA", style=line, linewidth=1, color=colFinal,transp=10)
p2=plot(showRibbons?emaF11:na, title="Fast Ribbon Upper MA", style=line, linewidth=1, color=colFinal,transp=10)
plot(showAvgs?emafast:na, title="Fast Ribbon Avg MA", style=circles,join=true, linewidth=1, color=gold,transp=10)

//
//fill(p1,p2,color=colFinal, transp=90)

//Slow EMA Plots
p3=plot(showRibbons?emaS1:na, title="Slow Ribbon Lower MA", style=line, linewidth=1, color=colFinal2,transp=10)
p4=plot(showRibbons?emaS16:na, title="Slow Ribbon Upper MA", style=line, linewidth=1, color=colFinal2,transp=10)
plot(showAvgs?emaslow:na, title="Slow Ribbon Avg MA", style=circles,join=true, linewidth=1, color=fuchsia,transp=10)
//
//fill(p3,p4, color=colFinal2, transp=90)

// Generate Buy Sell signals, 
buy = 0
sell=0
//
buy  := xup>=2 and sup>=2 and fup>=2 ? nz(buy[1])>0?buy[1]+1  :1 : 0
sell := xdn>=2 and sdn>=2 and fdn>=2 ? nz(sell[1])>0?sell[1]+1 :1 : 0

////////////////////////
//* 反测试周期选择器 *//
////////////////////////

testStartYear = input(2018, "Backtest Start Year",minval=1980)
testStartMonth = input(1, "Backtest Start Month",minval=1,maxval=12)
testStartDay = input(1, "Backtest Start Day",minval=1,maxval=31)
testPeriodStart = timestamp(testStartYear,testStartMonth,testStartDay,0,0)

testStopYear = 9999 //input(9999, "Backtest Stop Year",minval=1980)
testStopMonth = 12 // input(12, "Backtest Stop Month",minval=1,maxval=12)
testStopDay = 31 //input(31, "Backtest Stop Day",minval=1,maxval=31)
testPeriodStop = timestamp(testStopYear,testStopMonth,testStopDay,0,0)

testPeriod() => time >= testPeriodStart and time <= testPeriodStop ? true : false


///////////////
//* RSI策略 *//
///////////////

//指示器 1
lowerpc = lowest(low, 21)
upperpc = highest(high, 21)
midpc = avg(upperpc, lowerpc)

//指示器 2
ma = sma(close, 50)
petd = ema(close,13)
rangema = ema(tr, 50)
upperkc = ma + rangema * 0.25
lowerkc = ma - rangema * 0.25

//指示器 3
up = rma(max(change(close), 0), 5)
down = rma(-min(change(close), 0), 5)
rsi = down == 0 ? 100 : up == 0 ? 0 : 100 - (100 / (1 + up / down))

// PET-D
petdcolor = close > petd ? green : red
barcolor (petdcolor)


//Slope
SlopeL = midpc > midpc[5]
SlopeS = midpc < midpc[5]

//条件
CL = SlopeL == 1 and close > lowerkc and close < midpc and rsi < 35 
CS = SlopeS == 1 and close < upperkc and close > midpc and rsi > 65 

//Setup
RsiSL = CL == 1 and CL[1] != 1
RsiSS = CS == 1 and CS[1] != 1

/////////////////////
//* RSA抛物线指标 *//
/////////////////////
start = input(0.02)
increment = input(0.02)
maximum = input(0.2)

psar = sar(start, increment, maximum)

RSALE = false
RSASE = false
if (psar > high)
    RSALE = true
else
    RSALE = false

if (psar < low)
    RSASE = true
else
    RSASE = false


////////////////
//* 策略组件 *//
////////////////

AQUA = #00FFFFFF
BLUE = #0000FFFF
RED  = #FF0000FF
LIME = #00FF00FF
GRAY = #808080FF
DARKRED   = #8B0000FF
DARKGREEN = #006400FF
//
fastExit  = input(false,title="Use Opposite Trade as a Close Signal")
clrBars   = input(true,title="Colour Candles to Trade Order state")
orderType = input("Longs+Shorts",title="What type of Orders", options=["Longs+Shorts","LongsOnly","ShortsOnly","Flip"])
//
isLong   = (orderType != "ShortsOnly")
isShort  = (orderType != "LongsOnly")

//////////////////////////
//* 贸易状态引擎 *//
//////////////////////////

// 追踪当前贸易状态
longClose = false, longClose := nz(longClose[1],false)
shortClose = false, shortClose := nz(shortClose[1],false)
tradeState = 0, tradeState := nz(tradeState[1])
tradeState := tradeState==0 ?   buy==1 and (barstate.isconfirmed or barstate.ishistory) and isLong and not longClose and not shortClose? 1 :
                               sell==1  and (barstate.isconfirmed or barstate.ishistory) and isShort and not longClose and not shortClose? -1 : 
                          tradeState : tradeState
                          
////////////////////////////////////
//* 在这里设置入口和特殊出口条件 *//
////////////////////////////////////

//进入状态，当状态改变方向时。
longCondition  = false
shortCondition = false
//longCondition  := longCondition != true ? change(tradeState) and tradeState==1 : true
//shortCondition := shortCondition != true ? change(tradeState) and tradeState==-1 : true
longCondition  := change(tradeState) and tradeState==1
shortCondition := change(tradeState) and tradeState==-1
if orderType=="Flip"
    temp = longCondition
    longCondition  := shortCondition
    shortCondition := temp
//end if

// 卖出信号退出
longExitC  =  (emafast[1]<emaslow[1] and open<emaslow[1]) ? 1 : 0
shortExitC = (emafast[1]>emaslow[1] and open>emaslow[1]) ? 1 : 0

// change退出条件。
longExit = change(longExitC) and longExitC==1 and tradeState==1
shortExit = change(shortExitC) and shortExitC==1 and tradeState==-1

// -- debugs
//plotchar(tradeState,"tradeState at Event",location=location.bottom, color=#FF0000FF)
//plotchar(longCondition, title="longCondition",color=#FF0000FF)
//plotchar(shortCondition, title="shortCondition",color=#FF0000FF)
//plotchar(tradeState, title="tradeState",color=#006400FF)
// -- /debugs

////////////////////////////////
//======[ 交易入门价格 ]======//
////////////////////////////////

last_open_longCondition = na
last_open_shortCondition = na
last_open_longCondition := longCondition ? close : nz(last_open_longCondition[1])
last_open_shortCondition := shortCondition ? close : nz(last_open_shortCondition[1])

////////////////////////////
//======[ 位置状态 ]======//
////////////////////////////

in_longCondition  = tradeState == 1 
in_shortCondition = tradeState == -1

////////////////////////
//======[ 尾停 ]======//
////////////////////////

isTS = input(true, "Trailing Stop")
ts = input(3.0, "Trailing Stop (%)", minval=0,step=0.1, type=float) /100

last_high = na
last_low = na
last_high_short = na
last_low_long = na
last_high := not in_longCondition ? na : in_longCondition and (na(last_high[1]) or high_ > nz(last_high[1])) ? high_ : nz(last_high[1])
last_high_short := not in_shortCondition ? na : in_shortCondition and (na(last_high[1]) or high_ > nz(last_high[1])) ? high_ : nz(last_high[1])
last_low := not in_shortCondition ? na : in_shortCondition and (na(last_low[1]) or low_ < nz(last_low[1])) ? low_ : nz(last_low[1])
last_low_long := not in_longCondition ? na : in_longCondition and (na(last_low[1]) or low_ < nz(last_low[1])) ? low_ : nz(last_low[1])

long_ts =  isTS and in_longCondition and not na(last_high) and (low_ <= last_high - last_high * ts) //and (last_high >= last_open_longCondition + last_open_longCondition * tsi)
short_ts = isTS and in_shortCondition and not na(last_low) and (high_ >= last_low + last_low * ts) //and (last_low <= last_open_shortCondition - last_open_shortCondition * tsi)


////////////////////////
//======[ 获利 ]======//
////////////////////////

isTP = input(true, "Take Profit")
tp = input(3.0, "Take Profit (%)",minval=0,step=0.1,type=float) / 100
ttp = input(1.0, "Trailing Profit (%)",minval=0,step=0.1,type=float) / 100
ttp := ttp>tp ? tp : ttp

long_tp =  isTP and in_longCondition  and (last_high >= last_open_longCondition + last_open_longCondition * tp)   and (low_ <= last_high - last_high * ttp)
short_tp = isTP and in_shortCondition and (last_low <= last_open_shortCondition - last_open_shortCondition * tp) and (high_ >= last_low + last_low * ttp)

////////////////////////////
//======[ 停止损耗 ]======//
////////////////////////////

isSL = input(false, "Stop Loss")
sl = input(3.0, "Stop Loss (%)", minval=0,step=0.1, type=float) / 100
long_sl =  isSL and in_longCondition and (low_ <= last_open_longCondition - last_open_longCondition * sl)
short_sl = isSL and in_shortCondition and (high_ >= last_open_shortCondition + last_open_shortCondition * sl)

////////////////////////
//======[ 对峙 ]======//
////////////////////////

//注：短出口信号不重漆，无需用力，如果锥体继续进行。
long_sos = (fastExit or (not isTS and not isSL)) and longExit and in_longCondition
short_sos = (fastExit or (not isTS and not isSL)) and shortExit and in_shortCondition 

////////////////////////////
//======[ 关闭信号 ]======//
////////////////////////////

// 为所有不同的关闭条件创建一个单独的关闭，这里的所有条件都不重漆。
longClose :=  isLong and (long_tp or long_sl or long_ts or long_sos) and not longCondition
shortClose := isShort and (short_tp or short_sl or short_ts or short_sos) and not shortCondition
////////////////////////////
//======[ 情节色彩 ]======//
////////////////////////////

longCloseCol = na
shortCloseCol = na
longCloseCol := long_tp ? green : long_sl ? maroon : long_ts ? purple : long_sos ? orange :longCloseCol[1]
shortCloseCol := short_tp ? green : short_sl ? maroon : short_ts ? purple : short_sos ? orange : shortCloseCol[1]
//
tpColor = isTP and in_longCondition ? lime : isTP and in_shortCondition ? lime : na
slColor = isSL and in_longCondition ? red : isSL and in_shortCondition ? red : na


//////////////////////////////////
//======[ 策略图 ]======//
//////////////////////////////////

plot(isTS and in_longCondition?
     last_high - last_high * ts : na, "Long Trailing", fuchsia, style=2, linewidth=2,offset=1)
plot(isTP and in_longCondition and last_high < last_open_longCondition + last_open_longCondition * tp ? 
     last_open_longCondition + last_open_longCondition * tp : na, "Long TP Active", tpColor, style=3,join=false, linewidth=2,offset=1)
plot(isTP and in_longCondition and last_high >= last_open_longCondition +  last_open_longCondition * tp ? 
     last_high - last_high * ttp : na, "Long Trailing", black, style=2, linewidth=2,offset=1)
plot(isSL and in_longCondition and last_low_long > last_open_longCondition - last_open_longCondition * sl ? 
     last_open_longCondition - last_open_longCondition * sl : na, "Long SL", slColor, style=3,join=false, linewidth=2,offset=1)

plot(isTS and in_shortCondition?
     last_low + last_low * ts : na, "Short Trailing", fuchsia, style=2, linewidth=2,offset=1)
plot(isTP and in_shortCondition and last_low > last_open_shortCondition - last_open_shortCondition * tp ? 
     last_open_shortCondition - last_open_shortCondition * tp : na, "Short TP Active", tpColor, style=3,join=false, linewidth=2,offset=1)
plot(isTP and in_shortCondition and last_low <= last_open_shortCondition -  last_open_shortCondition * tp ? 
     last_low + last_low * ttp : na, "Short Trailing", black, style=2, linewidth=2,offset=1)
plot(isSL and in_shortCondition and last_high_short < last_open_shortCondition + last_open_shortCondition * sl ? 
     last_open_shortCondition + last_open_shortCondition * sl : na, "Short SL", slColor, style=3,join=false, linewidth=2,offset=1)

bclr = not clrBars ? na : tradeState==0 ? GRAY : 
                          in_longCondition ? close<last_open_longCondition? DARKGREEN : LIME :
                          in_shortCondition ? close>last_open_shortCondition? DARKRED : RED : GRAY
barcolor(bclr,title="Trade State Bar Colouring")


//////////////////////////////////
//======[ 战略进入与退出 ]======//
//////////////////////////////////

if testPeriod() and isLong
    strategy.entry("Long", 1, when=longCondition)
    strategy.close("Long", when=longClose )

if testPeriod() and isShort
    strategy.entry("Short", 0,  when=shortCondition)
    strategy.close("Short", when=shortClose )
    
// --- Debugs
//plotchar(longExit,title="longExit",location=location.bottom,color=na)
//plotchar(longCondition,title="longCondition",location=location.bottom,color=na)
//plotchar(in_longCondition,title="in_longCondition",location=location.bottom,color=na)
//plotchar(longClose,title="longClose",location=location.bottom,color=na,color=na)
//plotchar(buy,title="buy",location=location.bottom,color=na)
// --- /Debugs
//开单时改变背景
bgcolor( in_longCondition ? lime : na, transp=90)
bgcolor( in_shortCondition ? red : na, transp=90)

////////////////////////////
//======[ 重置变量 ]======//
////////////////////////////


if longClose or not in_longCondition
    last_high := na
    last_high_short := na
    
if shortClose or not in_shortCondition
    last_low := na
    last_low_long := na
    
if longClose or shortClose
    tradeState := 0
    in_longCondition := false
    in_shortCondition := false

    
//plotchar(tradeState,"tradeState at EOF",location=location.bottom, color=na)
// EOF

Mais.