Strategi tren kuantitatif yang menggabungkan beberapa faktor
Tanggal Pembuatan:
2024-01-12 11:09:40
Akhirnya memodifikasi:
2024-01-12 11:09:40
menyalin:
1
Jumlah klik:
572
ChaoZhang
1
fokus pada
1617
Pengikut
Ringkasan
Strategi ini mengintegrasikan beberapa faktor seperti volume transaksi, volatilitas, posisi penutupan, tren, dan lain-lain untuk mengidentifikasi peluang perdagangan, yang merupakan strategi kuantitatif multi-faktor yang khas.
Prinsip Strategi
Gagasan inti dari strategi ini adalah menggabungkan beberapa faktor, seperti penembusan yang tidak biasa dalam volume, posisi penutupan, dan amplitudo fluktuasi untuk mengidentifikasi titik jual beli.
Secara khusus, strategi ini menghitung volume transaksi rata-rata selama beberapa waktu terakhir, dan ketika volume transaksi dalam siklus saat ini terjadi, mungkin ada perubahan yang jelas. Selain itu, jika harga penutupan mendekati batas atas atau bawah dari amplitudo fluktuasi, itu berarti bahwa tren saat ini mungkin berbalik.
Untuk memverifikasi titik jual beli, strategi ini juga akan digabungkan dengan indikator amplitudo fluktuasi untuk menilai. Jika fluktuasi menembus rata-rata selama beberapa waktu terakhir, maka akan terbentuk kondisi pertama dari sinyal jual beli. Kemudian, jika harga penutupan siklus naik berada di bagian bawah dari amplitudo fluktuasi, dan volume transaksi meningkat, maka akan dihasilkan sinyal jual. Sebaliknya, jika harga penutupan siklus turun berada di bagian atas dari amplitudo fluktuasi, dan volume transaksi berkurang, maka akan dihasilkan sinyal beli.
Selain itu, strategi ini juga akan dikombinasikan dengan indikator garis rata untuk menilai tren keseluruhan, dan jika terjadi perubahan tren pada garis tengah dan panjang, ini juga akan berfungsi sebagai kondisi untuk menghasilkan sinyal beli dan jual.
Dengan menggabungkan berbagai indikator yang disebutkan di atas, strategi ini dapat memberikan penilaian yang komprehensif tentang waktu jual beli di pasar.
Keunggulan Strategis
Keuntungan utama dari strategi ini adalah mempertimbangkan berbagai faktor untuk membuat keputusan secara menyeluruh, sehingga hasilnya lebih dapat diandalkan. Secara khusus, ada beberapa keuntungan utama:
- Pertimbangannya adalah bahwa jika terjadi lonjakan lalu lintas yang tidak biasa, maka akan lebih awal untuk mendeteksi perubahan tren.
- Untuk mengevaluasi tren yang sebenarnya, perhatikan besarnya pergerakan dan posisi penutupan, dan jangan tertipu oleh kebisingan pasar jangka pendek.
- Strategi ini dapat digabungkan dengan garis rata-rata jangka menengah dan panjang untuk menentukan arah tren secara keseluruhan dan memastikan bahwa strategi tersebut berjalan sesuai dengan tren pasar utama.
- Keputusan komprehensif multi-faktor membuat strategi lebih stabil dan dapat diandalkan, menghindari kerugian besar.
Risiko Strategis
Strategi ini juga memiliki beberapa risiko yang perlu diperhatikan:
- Pertimbangan kombinasi multi faktor lebih rumit, dan parameter tuning lebih sulit.
- Tidak bisa sepenuhnya menghindari terjadinya transaksi yang tidak perlu karena tertipu oleh peretasan palsu.
- Jika salah menilai tren utama, efek operasional secara keseluruhan akan dikurangi.
- Pengaturan parameter volatilitas perlu disesuaikan dengan kondisi pasar yang berbeda.
Arah optimasi
Strategi ini juga memiliki beberapa optimasi utama:
- Mengoptimalkan parameter secara otomatis menggunakan metode seperti pembelajaran mesin.
- Bergabunglah dengan strategi manajemen risiko Stop Loss.
- Selain itu, ada beberapa faktor lain seperti aliran dana dan lain-lain yang dapat digunakan untuk menentukan tren utama.
- Desain beradaptasi dengan parameter fluktuasi.
Meringkaskan
Strategi ini mempertimbangkan berbagai faktor untuk mengidentifikasi peluang perdagangan. Keunggulan strategi adalah penilaian yang komprehensif, stabil dan dapat diandalkan. Risiko utama adalah kesalahan dalam menentukan parameter dan menilai tren utama. Di masa depan, parameter dapat dioptimalkan melalui pembelajaran mesin dan lain-lain.
Kode Sumber Strategi
/*backtest
start: 2023-12-12 00:00:00
end: 2024-01-11 00:00:00
period: 1h
basePeriod: 15m
exchanges: [{"eid":"Futures_Binance","currency":"BTC_USDT"}]
*/
//@version=3
strategy("volume spread analysis ", overlay=true)
volavg = sma(volume,40)
c= close
l=low
h=high
v=volume
volmean = stdev(volavg,30)
volupband3 = volavg + 3*volmean
volupband2 = volavg + 2*volmean
volupband1 = volavg + 1*volmean
voldnband1 = volavg -1*volmean
voldnband2 = volavg -2*volmean
midprice = (high+low)/2
spread = (high-low)
avgspread = sma(spread,80)
avgspreadbar = spread > avgspread
widerangebar = spread>(1.5*avgspread)
narrowrangebar = spread<(0.7*avgspread)
lowvolume = volume<volume[1] and volume<volume[2]
upbar = close>close[1]
downbar = close<close[1]
highvolume = volume>volume[1] and volume[1]>volume[2]
closefactor = close-low
clsposition = spread/closefactor
closeposition = iff(closefactor==0,avgspread,clsposition)
vb = volume>volavg or volume>volume[1]
upclose = close>=((spread*0.7)+low)// close is above 70% of the bar
downclose = close<=((spread*0.3)+low)// close is below the 30% of the bar
aboveclose = close>((spread*0.5)+low)// close is between 50% and 70% of the bar
belowclose = close<((spread*0.5)+low)// close is between 50% and 30% of the bar
midclose = close>((spread*0.3)+low) and c<((spread*0.7)+l)// close is between 30% and 70% of the bar
verylowclose = closeposition>4//close is below 25% of the bar
veryhighclose = closeposition<1.35// close is above 80% of the bar
closepos = iff(close<=((spread*0.2)+low),1,iff(close<=((spread*0.4)+low),2,iff(close<=((spread*0.6)+low),3,iff(close<=((spread*0.8)+low),4,5))))
// 1 = downclose, 2 = belowclose, 3 = midclose, 4 = aboveclose, 5 = upclose
volpos = iff(volume>volavg*2,1,iff(volume>volavg*1.3,2,iff(volume>volavg,3,iff(volume<volavg and volume>volavg*0.7,4,5))))
//// 1 = very high, 2 = high, 3 = above average, 4 = less than average, 5 = low
freshgndhi = close > highestbars(h,5)
freshgndlo = close < lowestbars(l,5)
//========================trend estimation =========================
//jtrend=sma(close,5)
//trendlongterm = linreg(jtrend,40)
//trendmediumterm = linreg(jtrend,10)
//trendshortterm = linreg(jtrend,3)
//tls=linreg(jtrend,3)
minperiodsrwist = input(title="short term min periods", defval=2, minval=1)
maxperiodsrwist = input(title="short term max periods", defval=8, minval=1)
minperiodsrwilt = input(title="long term min periods", defval=10, minval=1)
maxperiodsrwilt = input(title="long term max periods", defval=40, minval=1)
rwhmins = (high - nz(low[minperiodsrwist])) / (atr(minperiodsrwist) * sqrt(minperiodsrwist))
rwhmaxs = (high - nz(low[maxperiodsrwist])) / (atr(maxperiodsrwist) * sqrt(maxperiodsrwist))
rwhs = max( rwhmins, rwhmaxs )
rwlmins = (nz(high[minperiodsrwist]) - low) / (atr(minperiodsrwist) * sqrt(minperiodsrwist))
rwlmaxs = (nz(high[maxperiodsrwist]) - low) / (atr(maxperiodsrwist) * sqrt(maxperiodsrwist))
rwls = max( rwlmins, rwlmaxs )
rwhminl = (high - nz(low[minperiodsrwilt])) / (atr(minperiodsrwilt) * sqrt(minperiodsrwilt))
rwhmaxl = (high - nz(low[maxperiodsrwilt])) / (atr(maxperiodsrwilt) * sqrt(maxperiodsrwilt))
rwhl = max( rwhminl, rwhmaxl )
rwlminl = (nz(high[minperiodsrwilt]) - low) / (atr(minperiodsrwilt) * sqrt(minperiodsrwilt))
rwlmaxl = (nz(high[maxperiodsrwilt]) - low) / (atr(maxperiodsrwilt) * sqrt(maxperiodsrwilt))
rwll = max( rwlminl, rwlmaxl )
ground = rwhs
sky = rwls
j = rwhs-rwls
k = rwhl-rwll
j2 = rwhl
k2 = rwll
ja = cross(j,1)
jb = cross(1,j)
jc = cross(-1,j)
jd = cross(j,-1)
j2a = cross(j2,1)
j2b = cross(1,j2)
k2a = cross(k2,1)
k2b = cross(1,k2)
upmajoron = j > 1 and ja[1]
upmajoroff = j < 1 and jb[1]
upminoron = j2 > 1 and j2a[1]
upminoroff = j2 < 1 and j2b[1]
dnmajoron = j < -1 and jc[1]
dnmajoroff = j > -1 and jd[1]
dnminoron = k2 > 1 and k2a[1]
dnminoroff = k2 < 1 and k2b[1]
upimd = iff(ground > 1, 1,0)
dnimd = iff(sky > 1, 1, 0)
upmajor = iff(j>1,1,iff(j<(-1),-1,0))
upminor = iff(j2>1,1,-1)
dnminor = iff(k2>1,1,-1)
//======================================================================|
Buy_stop = lowest(low[1],5) - atr(20)[1]
plot(Buy_stop, color=red, title="buy_stoploss")
Sell_stop = highest(high[1],5) + atr(20)[1]
plot(Sell_stop, color=green, title="sell_stoploss")
//======================================================================|
//upthrustbar = widerangebar and downclose and upimd==1 and high>high[1] //wrb and uhs and fresh ground
nut = widerangebar and downclose and freshgndhi and highvolume // new signal
bc = widerangebar and aboveclose and volume == highest(volume,60) and upmajor==1 // new signal
upthrustbar = widerangebar and (closepos==1 or closepos==2) and upminor>0 and high>high[1] and (upimd>0or upmajor>0) and volpos <4// after minor up trend
upthrustbartrue = widerangebar and closepos==1 and upmajor>0 and high>high[1] and volpos <4//occurs after a major uptrend
upthrustcond1 = upthrustbar[1] and downbar and not narrowrangebar
upthrustcond2 = upthrustbar[1] and downbar and volpos == 2
upthrustcond3 = upthrustbar and volpos ==1
toprevbar = volume[1]>volavg and upbar[1] and widerangebar[1] and downbar and downclose and widerangebar and upmajor>0 and high==highest(high,10)
pseudoupthrust = upbar[1] and high>high[1] and volume[1]>1.5*volavg and downbar and downclose and not upthrustbar
pseudoutcond = pseudoupthrust[1] and downbar and downclose and not upthrustbar
trendchange = upbar[1] and high==highest(high,5) and downbar and (downclose or midclose) and volume>volavg and upmajor>0 and upimd>0 and not widerangebar and not pseudoupthrust
nodemandbarut = upbar and narrowrangebar and lowvolume and closepos> 3 and ((upminor>0 and upimd>0)or (upminor<0 and upminor>0))//in a up market
nodemandbardt = upbar and narrowrangebar and lowvolume and closepos> 3 and (upminor<=0or upimd<=0)// in a down or sidewayss market
nosupplybar = downbar and narrowrangebar and lowvolume and closepos<3 and ((upminor<1 and upimd<1)or (upminor>0 and upimd<1))
lowvoltest = low==lowest(low,5) and upclose and lowvolume//lowvolume and l<low[1] and upclose
lowvoltest1 = low==lowest(low,5) and volume<volavg and low<low[1] and upclose and upminor>0 and upmajor>0// and widerangebar
lowvoltest2 = lowvoltest[1] and upbar and upclose
sellcond1 = (upthrustcond1 or upthrustcond2 or upthrustcond3)
sellcond2 = sellcond1[1]==0
sellcond = sellcond1 and sellcond2
strengthdown0 = upmajor<0 and volpos<4 and downbar[1] and upbar and closepos>3 and upminor<0 and upimd<=0// strength after a long down trend
strengthdown = volpos<4 and downbar[1] and upbar and closepos>3 and upimd<=00 and upminor<0// strength after a down trend
strengthdown1 = upmajor<0 and volume>(volavg*1.5) and downbar[1] and upbar and closepos>3 and upminor<0 and upimd<=0//strength after downtrend . high volume
strengthdown2 = upimd<=0 and volume[1]<volavg and upbar and veryhighclose and volpos<4
buycond1 = strengthdown or strengthdown1
buycond = upbar and buycond1[1]
stopvolume = low==lowest(low,5) and (upclose or midclose) and v>1.5*volavg and upmajor<0
revupthrust = upmajor<0 and upbar and upclose and volume>volume[1] and volume>volavg and widerangebar and downbar[1] and downclose[1] and upminor<0
effortup = high>high[1] and low>low[1] and close>close[1] and close>=((high-low)*0.7+low) and spread>avgspread and volpos<4//and open<=((high-low)*0.3+low)
effortupfail = effortup[1] and (upthrustbar or upthrustcond1 or upthrustcond2 or upthrustcond3 or (downbar and avgspreadbar))
effortdown = high<high[1] and low<low[1] and close<close[1] and close<=((high-low)*0.25+low) and widerangebar and volume>volume[1]//o>=((high-low)*0.75+
effortdownfail = effortdown[1] and ((upbar and avgspreadbar)or revupthrust or buycond1)
upflag = (sellcond or buycond or effortup or effortupfail or stopvolume or effortdown or effortdownfail or revupthrust or nodemandbardt or nodemandbarut or nosupplybar or lowvoltest or lowvoltest1 or lowvoltest2 or bc)
bullbar = (volume>volavg or volume>volume[1]) and closeposition <2 and upbar and not upflag
bearbar = vb and downclose and downbar and spread>avgspread and not upflag
buy = (upbar and revupthrust[1])or lowvoltest2
burely = strengthdown1 and stopvolume[1]or (upbar and revupthrust[1])or lowvoltest2
//buy = effortup and lowvoltest2[1]
//sell = upthrustbartrue
sell = effortup[1] and effortupfail and upthrustcond3 and upthrustbartrue and toprevbar
strategy.entry("simpleBuy", strategy.long, when= (upbar and revupthrust[1])or lowvoltest2 )
strategy.close("simpleBuy",when=upthrustbartrue )
//strategy.entry("simpleSell", strategy.short,when= upthrustbartrue )
//strategy.close("simpleSell",when= (upbar and revupthrust[1])or lowvoltest2)
//|============================================================================================|
//data = close >= open
//plotshape(true, style=shape.flag, color=data ? green : red)
plotshape((upthrustbar or upthrustbartrue) ,title="upthrustbaro" ,style=shape.arrowdown ,size=size.huge,color=red )
//plotshape(toprevbar ,title="toprevbar" ,style=shape.flag ,size=size.small,color=blue )
//plotshape((pseudoupthrust) ,title="(pseudoupthrus" ,style=shape.circle ,size=size.small,color=blue )
//plotshape((upthrustcond1 or upthrustcond2) ,title="upthrustcond1" ,style=shape.triangleup ,size=size.small,color=red )
plotshape(trendchange ,title="trendchange" ,style=shape.xcross ,size=size.small,color=red )
//plotshape((nodemandbardt) ,title="(nodemandbardt" ,style=shape.square ,size=size.small,color=orange )
//plotshape(nosupplybar ,title="nosupplybar" ,style=shape.cross ,size=size.small,color=blue)
plotshape(revupthrust ,title="revupthrust" ,style=shape.arrowup ,size=size.huge,color=green )
//plotshape((upthrustbar or upthrustbartrue) ,title="upthrustbaro" ,style=shape.cross ,size=size.small,color=red )
//plotshape((upthrustcond1 or upthrustcond2) ,title="upthrustcond1" ,style=shape.triangledown ,size=size.small,color=white )
//plotshape((pseudoupthrust) ,title="(pseudoupthrus" ,style=shape.arrowup ,size=size.small,color=blue )
//plotshape(nodemandbarut ,title="nodemandbarut" ,style=shape.labelup ,size=size.small,color=orange )
//plotshape(nodemandbarut ,title="nodemandbarut" ,style=shape.labeldown ,size=size.small,color=yellow )
//plotshape(nodemandbardt ,title="nodemandbardt" ,style=shape.diamond ,size=size.small,color=yellow )
//plotshape(nosupplybar ,title="nosupplybar" ,style=shape.xcross ,size=size.small,color=blue )
plotshape(lowvoltest ,title="lowvoltest" ,style=shape.triangleup ,size=size.small,color=blue )
//plotshape(lowvoltest2 ,title="lowvoltest2" ,style=shape.triangledown ,size=size.small,color=yellow )
//plotshape(strengthdown ,title="strengthdown" ,style=shape.flag ,size=size.small,color=green)
//plotshape(strengthdown ,title="strengthdown" ,style=shape.circle ,size=size.small,color=lime )
//plotshape(strengthdown2 ,title="strengthdown2" ,style=shape.arrowup ,size=size.small,color=silver )
//plotshape(strengthdown2 ,title="strengthdown2" ,style=shape.arrowdown ,size=size.small,color=red )
//plotshape(stopvolume ,title="stopvolume" ,style=shape.labelup ,size=size.small,color=green )
//plotshape(stopvolume ,title="stopvolume" ,style=shape.labeldown ,size=size.small,color=yellow )
plotshape(effortup ,title="effortup" ,style=shape.diamond ,size=size.small,color=lime )
plotshape(effortupfail ,title="effortupfail" ,style=shape.xcross ,size=size.small,color=blue )
//plotshape(effortupfail ,title="effortupfail" ,style=shape.cross ,size=size.small,color=white )
plotshape(effortdown ,title="effortdown" ,style=shape.triangledown ,size=size.small,color=red )
plotshape(effortdownfail ,title="effortdownfail" ,style=shape.xcross ,size=size.small,color=green )
//plotshape(effortdownfail ,title="effortdownfail" ,style=shape.flag ,size=size.small,color=white )
//plotshape(buycond ,title="buycond" ,style=shape.circle ,size=size.small,color=green )
//plotshape(sellcond ,title="sellcond" ,style=shape.arrowup ,size=size.small,color=orange )
//plotshape((nut) ,title="(nut)" ,style=shape.arrowdown ,size=size.small,color=lime )
//plotshape((bc ) ,title="(bc" ,style=shape.labelup ,size=size.small,color=red )
//plotshape(buy ,title="buy" ,style=shape.labeldown ,size=size.small,color=white )