Strategi Crossover Multi-Periode Rata-rata Bergerak
ChaoZhang
Ringkasan
Strategi ini didasarkan pada moving average crossover system, dengan moving average golden crossover dan death crossover dengan siklus yang berbeda untuk menilai waktu untuk membeli dan menjual. Strategi ini juga menggabungkan fitur trend tracking seperti stop loss, stop loss, dan stop loss untuk mengunci keuntungan dan menghindari risiko.
Prinsip Strategi
Strategi ini menggunakan dua set rata-rata bergerak, yaitu garis cepat dan garis lambat. Garis cepat memiliki siklus yang lebih pendek, yang mewakili tren jangka pendek; Garis lambat memiliki siklus yang lebih panjang, yang mewakili tren jangka panjang. Ketika garis cepat menerobos garis lambat dari arah bawah, ia menghasilkan golden cross, yang menunjukkan bahwa pasar telah berbalik dari bear ke bull, dan melakukan lebih banyak; Ketika garis cepat jatuh dari arah atas ke bawah dan melanggar garis lambat, ia membentuk death cross, yang menunjukkan bahwa pasar telah berbalik dari bull ke bear, dan melakukan kosong.
Dalam kode, indikator garis cepat adalah ma1 dan garis lambat adalah ma2. Ma1 dan ma2 dapat memilih berbagai jenis rata-rata bergerak, seperti SMA, EMA, dan lain-lain, dan dapat mengatur parameter periode yang berbeda. Ma1 mewakili tren jangka pendek, dengan periode yang lebih pendek; ma2 mewakili tren jangka panjang, dengan periode yang lebih panjang.
Ketika ma1 Gold Fork ma2 , menghasilkan sinyal panjang, melakukan lebih banyak; Ketika ma1 Dead Fork ma2 , menghasilkan sinyal pendek, melakukan short. Saat perdagangan nyata, dapat dikombinasikan dengan fitur trend tracking stop loss, stop loss, stop loss, dan lainnya, untuk mengunci keuntungan dan menghindari risiko.
Analisis Keunggulan
Strategi ini memiliki keuntungan sebagai berikut:
Strategi ini sederhana dan jelas, mudah dipahami dan diterapkan.
Fleksibel memilih moving average dari berbagai jenis dan parameter yang sesuai dengan kondisi pasar yang berbeda.
Desain multi-siklus, menangkap tren jangka pendek dan jangka panjang untuk menghindari kesalahan.
Kondisi pembukaan posisi dapat disesuaikan, dan frekuensi transaksi dikontrol secara ketat.
Anda dapat mengatur kondisi stop loss dan stop loss untuk mengontrol risiko.
Anda dapat menambahkan trend tracking, stop loss, dan profit tracking.
Parameter yang dapat dioptimalkan untuk membuat strategi lebih stabil dan dapat diandalkan.
Analisis risiko
Strategi ini juga memiliki risiko sebagai berikut:
Jika harga bergerak di atas rata-rata dua kali lipat, maka harga akan berbalik pada saat yang tepat.
Periode rata-rata bergerak yang tidak tepat dapat menghasilkan lebih banyak sinyal palsu.
Kejadian yang tak terduga menyebabkan terbaliknya serangan dengan cepat, dan menghentikan risiko terjatuh.
Dalam situasi tren, kemungkinan besar harga akan tetap berada di sisi garis rata-rata dalam jangka panjang.
Optimasi parameter yang tidak tepat, mungkin over-optimasi untuk jangka waktu tertentu.
Langkah-langkah manajemen risiko yang sesuai:
Dalam kombinasi dengan indikator lain, sinyal penyaringan untuk menghindari penembusan palsu.
Pengaturan siklus harus mengikuti prinsip perdagangan tren, menguji parameter optimasi.
Pengendalian risiko harus dilakukan dengan hati-hati dan pengaturan posisi stop loss harus masuk akal.
Anda harus menunggu dengan sabar.
Parameter pengujian kebugaran dalam berbagai lingkungan pasar.
Arah optimasi
Strategi ini dapat dioptimalkan dengan:
Uji lebih banyak jenis rata-rata bergerak, seperti rata-rata bergerak berbobot dan lain-lain.
Menambahkan siklus dinamis berdasarkan volatilitas, menyesuaikan parameter rata-rata sesuai dengan perubahan pasar.
Syarat masuk strategi dapat ditambahkan waktu atau filter dasar, untuk mengurangi tingkat kesalahan transaksi.
Kondisi keluar dapat mengatur stop loss dinamis, menyesuaikan stop loss sesuai dengan volatilitas pasar.
Anda dapat membuat sistem optimasi parameter untuk melakukan retrospeksi historis dan penyesuaian parameter dari strategi.
Menambahkan algoritma pembelajaran mesin, menggunakan AI untuk mengoptimalkan parameter dan memfilter sinyal.
Meringkaskan
Strategi bergerak rata-rata melintasi multi-siklus strategi keseluruhan ide yang jelas dan mudah dimengerti, dengan cepat dan lambat rata-rata melintasi untuk menangkap tren, adalah strategi yang lebih klasik dari trend tracking. Strategi dapat mencapai keuntungan yang stabil dengan memilih parameter yang tepat, mengoptimalkan masuk dan keluar logika, kontrol risiko yang ketat.
/*backtest
start: 2023-09-08 00:00:00
end: 2023-10-08 00:00:00
period: 4h
basePeriod: 15m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/
//@version=3
// The majority of this script I took from the Autoview website. There are some typos in the original that I've fixed, some things I've added, things I will add, and I'm tired pulling my strategy code out and uploading this to pastebin for people.
// DISCLAIMER: I am not a financial advisor, this is not financial advice, do not use this code without first doing your own research, etc, etc, it's not my fault when you lose your house.
strategy("Moving Averages Cross - MTF - Strategy", "MA Cross", overlay=true, pyramiding=0, initial_capital=100000, currency=currency.USD, default_qty_type = strategy.percent_of_equity, default_qty_value = 100, commission_type=strategy.commission.percent, commission_value=0.1)
bgcolor ( color=black, transp=40, title='Blackground', editable=true)
///////////////////////////////////////////////
//* Backtesting Period Selector | Component *//
///////////////////////////////////////////////
//* https://www.tradingview.com/script/eCC1cvxQ-Backtesting-Period-Selector-Component *//
//* https://www.tradingview.com/u/pbergden/ *//
//* Modifications made *//
testStartYear = input(2018, "Backtest Start Year")
testStartMonth = input(1, "Backtest Start Month")
testStartDay = input(1, "Backtest Start Day")
testPeriodStart = timestamp(testStartYear,testStartMonth,testStartDay,00,00)
testStopYear = input(9999, "Backtest Stop Year")
testStopMonth = input(12, "Backtest Stop Month")
testStopDay = input(31, "Backtest Stop Day")
testPeriodStop = timestamp(testStopYear,testStopMonth,testStopDay,0,0)
testPeriod() => true
/////////////////////////////////////
//* Put your strategy logic below *//
/////////////////////////////////////
sp1 = input("----", title="--------Moving Average 1----------", options=["----"])
maUseRes1 = input(defval = false, title = "Use Different Resolution?")
//maReso1 = input(defval = "60", title = "Set Resolution", type = resolution)
maReso1 = input(defval='60', title = "Set Resolution Minutes")
maType1 = input("EMA", title="MA", options=["SMA", "EMA", "DEMA", "TEMA", "WMA", "VWMA", "SMMA", "Hull", "LSMA", "ALMA"])
maSource1 = input(defval = close, title = "Source")
maLength1 = input(defval = 15, title = "Period", minval = 1)
lsmaOffset1 = input(defval = 1, title = "Least Squares (LSMA) Only - Offset Value", minval = 0)
almaOffset1 = input(defval = 0.85, title = "Arnaud Legoux (ALMA) Only - Offset Value", minval = 0, step = 0.01)
almaSigma1 = input(defval = 6, title = "Arnaud Legoux (ALMA) Only - Sigma Value", minval = 0)
sp2 = input("----", title="--------Moving Average 2----------", options=["----"])
maUseRes2 = input(defval = false, title = "Use Different Resolution?")
//maReso2 = input(defval = "60", title = "Set Resolution", type = resolution)
maReso2 = input(defval='60', title = "Set Resolution Minutes")
maType2 = input("EMA", title="MA", options=["SMA", "EMA", "DEMA", "TEMA", "WMA", "VWMA", "SMMA", "Hull", "LSMA", "ALMA"])
maSource2 = input(defval = close, title = "Source")
maLength2 = input(defval = 30, title = "Period", minval = 1)
lsmaOffset2 = input(defval = 1, title = "Least Squares (LSMA) Only - Offset Value", minval = 0)
almaOffset2 = input(defval = 0.85, title = "Arnaud Legoux (ALMA) Only - Offset Value", minval = 0, step = 0.01)
almaSigma2 = input(defval = 6, title = "Arnaud Legoux (ALMA) Only - Sigma Value", minval = 0)
//Function from @JayRogers thank you man awesome work
variant(type, src, len, lsmaOffset, almaOffset, almaSigma) =>
v1 = sma(src, len) // Simple
v2 = ema(src, len) // Exponential
v3 = 2 * v2 - ema(v2, len) // Double Exponential
v4 = 3 * (v2 - ema(v2, len)) + ema(ema(v2, len), len) // Triple Exponential
v5 = wma(src, len) // Weighted
v6 = vwma(src, len) // Volume Weighted
v7 = na(v5[1]) ? sma(src, len) : (v5[1] * (len - 1) + src) / len // Smoothed
v8 = wma(2 * wma(src, len / 2) - wma(src, len), round(sqrt(len))) // Hull
v9 = linreg(src, len, lsmaOffset) // Least Squares
v10 = alma(src, len, almaOffset, almaSigma) // Arnaud Legoux
type=="EMA"?v2 : type=="DEMA"?v3 : type=="TEMA"?v4 : type=="WMA"?v5 : type=="VWMA"?v6 : type=="SMMA"?v7 : type=="Hull"?v8 : type=="LSMA"?v9 : type=="ALMA"?v10 : v1
//Different resolution function
reso(exp, res, use) => use ? security(tickerid, res, exp) : exp
ma1 = reso(variant(maType1, maSource1, maLength1, lsmaOffset1, almaOffset1, almaSigma1), maReso1, maUseRes1)
ma2 = reso(variant(maType2, maSource2, maLength2, lsmaOffset2, almaOffset2, almaSigma2), maReso2, maUseRes2)
plotma1 = plot(ma1, color=green, tranps=50, linewidth = 2 )
plotma2 = plot(ma2, color=red, tranps=50, linewidth = 2 )
// Long/Short Logic
longLogic = crossover(ma1,ma2) ? 1 : 0
shortLogic = crossunder(ma1,ma2) ? 1 : 0
//////////////////////////
//* Strategy Component *//
//////////////////////////
isLong = input(false, "Longs Only")
isShort = input(false, "Shorts Only")
isFlip = input(false, "Flip the Opens")
long = longLogic
short = shortLogic
if isFlip
long := shortLogic
short := longLogic
else
long := longLogic
short := shortLogic
if isLong
long := long
short := na
if isShort
long := na
short := short
////////////////////////////////
//======[ Signal Count ]======//
////////////////////////////////
sectionLongs = 0
sectionLongs := nz(sectionLongs[1])
sectionShorts = 0
sectionShorts := nz(sectionShorts[1])
if long
sectionLongs := sectionLongs + 1
sectionShorts := 0
if short
sectionLongs := 0
sectionShorts := sectionShorts + 1
//////////////////////////////
//======[ Pyramiding ]======//
//////////////////////////////
pyrl = input(1, "Pyramiding less than") // If your count is less than this number
pyre = input(0, "Pyramiding equal to") // If your count is equal to this number
pyrg = input(1000000, "Pyramiding greater than") // If your count is greater than this number
longCondition = long and sectionLongs <= pyrl or long and sectionLongs >= pyrg or long and sectionLongs == pyre ? 1 : 0
shortCondition = short and sectionShorts <= pyrl or short and sectionShorts >= pyrg or short and sectionShorts == pyre ? 1 : 0
////////////////////////////////
//======[ Entry Prices ]======//
////////////////////////////////
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])
////////////////////////////////////
//======[ Open Order Count ]======//
////////////////////////////////////
sectionLongConditions = 0
sectionLongConditions := nz(sectionLongConditions[1])
sectionShortConditions = 0
sectionShortConditions := nz(sectionShortConditions[1])
if longCondition
sectionLongConditions := sectionLongConditions + 1
sectionShortConditions := 0
if shortCondition
sectionLongConditions := 0
sectionShortConditions := sectionShortConditions + 1
///////////////////////////////////////////////
//======[ Position Check (long/short) ]======//
///////////////////////////////////////////////
last_longCondition = na
last_shortCondition = na
last_longCondition := longCondition ? time : nz(last_longCondition[1])
last_shortCondition := shortCondition ? time : nz(last_shortCondition[1])
in_longCondition = last_longCondition > last_shortCondition
in_shortCondition = last_shortCondition > last_longCondition
/////////////////////////////////////
//======[ Position Averages ]======//
/////////////////////////////////////
totalLongs = 0.0
totalLongs := nz(totalLongs[1])
totalShorts = 0.0
totalShorts := nz(totalShorts[1])
averageLongs = 0.0
averageLongs := nz(averageLongs[1])
averageShorts = 0.0
averageShorts := nz(averageShorts[1])
if longCondition
totalLongs := totalLongs + last_open_longCondition
totalShorts := 0.0
if shortCondition
totalLongs := 0.0
totalShorts := totalShorts + last_open_shortCondition
averageLongs := totalLongs / sectionLongConditions
averageShorts := totalShorts / sectionShortConditions
/////////////////////////////////
//======[ Trailing Stop ]======//
/////////////////////////////////
isTS = input(false, "Trailing Stop")
tsi = input(1000, "Activate Trailing Stop Price (%). Divided by 100 (1 = 0.01%)") / 100
ts = input(575, "Trailing Stop (%). Divided by 100 (1 = 0.01%)") / 100
last_high = na
last_low = na
last_high_short = na
last_low_short = 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_short := 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 not na(last_high) and low <= last_high - last_high / 100 * ts and longCondition == 0 and last_high >= averageLongs + averageLongs / 100 * tsi
short_ts = isTS and not na(last_low) and high >= last_low + last_low / 100 * ts and shortCondition == 0 and last_low <= averageShorts - averageShorts/ 100 * tsi
///////////////////////////////
//======[ Take Profit ]======//
///////////////////////////////
isTP = input(false, "Take Profit")
tp = input(300, "Take Profit (%). Divided by 100 (1 = 0.01%)") / 100
long_tp = isTP and close > averageLongs + averageLongs / 100 * tp and not longCondition
short_tp = isTP and close < averageShorts - averageShorts / 100 * tp and not shortCondition
/////////////////////////////
//======[ Stop Loss ]======//
/////////////////////////////
isSL = input(false, "Stop Loss")
sl = input(575, "Stop Loss (%). Divided by 100 (1 = 0.01%)") / 100
long_sl = isSL and close < averageLongs - averageLongs / 100 * sl and longCondition == 0
short_sl = isSL and close > averageShorts + averageShorts / 100 * sl and shortCondition == 0
/////////////////////////////////
//======[ Close Signals ]======//
/////////////////////////////////
longClose = long_tp or long_sl or long_ts ? 1 : 0
shortClose = short_tp or short_sl or short_ts ? 1: 0
///////////////////////////////
//======[ Plot Colors ]======//
///////////////////////////////
longCloseCol = na
shortCloseCol = na
longCloseCol := long_tp ? purple : long_sl ? maroon : long_ts ? blue : longCloseCol[1]
shortCloseCol := short_tp ? purple : short_sl ? maroon : short_ts ? blue : shortCloseCol[1]
tpColor = isTP and in_longCondition ? purple : isTP and in_shortCondition ? purple : white
slColor = isSL and in_longCondition ? red : isSL and in_shortCondition ? red : white
//////////////////////////////////
//======[ Strategy Plots ]======//
//////////////////////////////////
// Comment out these lines to use alerts
plot(isTS and in_longCondition ? averageLongs + averageLongs / 100 * tsi : na, "Long Trailing Activate", blue, style=3, linewidth=2)
plot(isTS and in_longCondition and last_high >= averageLongs + averageLongs / 100 * tsi ? last_high - last_high / 100 * ts : na, "Long Trailing", fuchsia, style=2, linewidth=3)
plot(isTS and in_shortCondition ? averageShorts - averageShorts/ 100 * tsi : na, "Short Trailing Activate", blue, style=3, linewidth=2)
plot(isTS and in_shortCondition and last_low <= averageShorts - averageShorts/ 100 * tsi ? last_low + last_low / 100 * ts : na, "Short Trailing", fuchsia, style=2, linewidth=3)
plot(isTP and in_longCondition and last_high < averageLongs + averageLongs / 100 * tp ? averageLongs + averageLongs / 100 * tp : na, "Long TP", tpColor, style=3, linewidth=2)
plot(isTP and in_shortCondition and last_low > averageShorts - averageShorts / 100 * tp ? averageShorts - averageShorts / 100 * tp : na, "Short TP", tpColor, style=3, linewidth=2)
plot(isSL and in_longCondition and last_low_short > averageLongs - averageLongs / 100 * sl ? averageLongs - averageLongs / 100 * sl : na, "Long SL", slColor, style=3, linewidth=2)
plot(isSL and in_shortCondition and last_high_short < averageShorts + averageShorts / 100 * sl ? averageShorts + averageShorts / 100 * sl : na, "Short SL", slColor, style=3, linewidth=2)
///////////////////////////////
//======[ Alert Plots ]======//
///////////////////////////////
// Uncomment to use Alerts, or the new Signal Plots, but not both
// Old Signal Plots
//plot(longCondition, "Long", green)
//plot(shortCondition, "Short", red)
//plot(longClose, "Long Close", longCloseCol)
//plot(shortClose, "Short Close", shortCloseCol)
// Uncomment for your alerts
//alertcondition(condition=longCondition, title="Long", message="")
//alertcondition(condition=shortCondition, title="Short", message="")
//alertcondition(condition=longClose, title="Long Close", message="")
//alertcondition(condition=shortClose, title="Short Close", message="")
///////////////////////////////////
//======[ Reset Variables ]======//
///////////////////////////////////
if longClose or not in_longCondition
averageLongs := 0
totalLongs := 0.0
sectionLongs := 0
sectionLongConditions := 0
if shortClose or not in_shortCondition
averageShorts := 0
totalShorts := 0.0
sectionShorts := 0
sectionShortConditions := 0
////////////////////////////////////////////
//======[ Strategy Entry and Exits ]======//
////////////////////////////////////////////
// Comment out to use alerts
if testPeriod()
strategy.entry("Long", 1, when=longCondition)
strategy.entry("Short", 0, when=shortCondition)
strategy.close("Long", when=longClose)
strategy.close("Short", when=shortClose)