本页源码来自 TradingView 公开发布的开源脚本，版权归原作者所有， 请遵循其原始许可（Pine 脚本常见 CC BY-NC-SA / MPL-2.0 / MIT）。 本项目仅用于研究检索与许可范围内的移植。
// This work is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International
// https://creativecommons.org/licenses/by-nc-sa/4.0/
// © Zeiierman {
//@version=6
indicator('SuperTrend Cluster (Zeiierman)', max_labels_count = 200, overlay = true, max_bars_back = 2000, behind_chart = false)
//}
// ~~ Tooltips {
var string t1 = "Minimum weighted agreement required for the bullish or bearish cluster to become valid. Higher values demand stronger alignment across the SuperTrend set."
var string t2 = "Selects which one of the five SuperTrend members is used as the base reference for flip markers, label placement, and final direction alignment."
var string t3 = "Colors the candles and bars using the live cluster strength gradient. When disabled, chart candles keep their default chart colors."
var string t4 = "Shows or hides the Bull Cluster and Bear Cluster labels when the selected base SuperTrend flips."
var string t5 = "Shows or hides the small base SuperTrend flip markers plotted at the selected base SuperTrend line."
var string t6 = "Main bullish color used for bullish trend lines, bullish labels, bullish markers, and bullish candle coloring."
var string t7 = "Main bearish color used for bearish trend lines, bearish labels, bearish markers, and bearish candle coloring."
var string t8 = "Neutral midpoint color used by the bar and candle gradient when bullish and bearish cluster pressure is balanced."
var string t9 = "ATR length for SuperTrend 1. Lower values react faster to price changes, while higher values make this member slower and smoother."
var string t10 = "ATR multiplier for SuperTrend 1. Higher values place the band farther from price and reduce sensitivity."
var string t11 = "Smoothing method applied to the source before SuperTrend 1 is calculated."
var string t12 = "Length of the smoothing used for SuperTrend 1. Higher values smooth more but add lag."
var string t13 = "Relative influence of SuperTrend 1 inside the weighted cluster. Higher values make this member contribute more to the final consensus."
var string t14 = "ATR length for SuperTrend 2. Lower values react faster to price changes, while higher values make this member slower and smoother."
var string t15 = "ATR multiplier for SuperTrend 2. Higher values place the band farther from price and reduce sensitivity."
var string t16 = "Smoothing method applied to the source before SuperTrend 2 is calculated."
var string t17 = "Length of the smoothing used for SuperTrend 2. Higher values smooth more but add lag."
var string t18 = "Relative influence of SuperTrend 2 inside the weighted cluster. Higher values make this member contribute more to the final consensus."
var string t19 = "ATR length for SuperTrend 3. Lower values react faster to price changes, while higher values make this member slower and smoother."
var string t20 = "ATR multiplier for SuperTrend 3. Higher values place the band farther from price and reduce sensitivity."
var string t21 = "Smoothing method applied to the source before SuperTrend 3 is calculated."
var string t22 = "Length of the smoothing used for SuperTrend 3. Higher values smooth more but add lag."
var string t23 = "Relative influence of SuperTrend 3 inside the weighted cluster. Higher values make this member contribute more to the final consensus."
var string t24 = "ATR length for SuperTrend 4. Lower values react faster to price changes, while higher values make this member slower and smoother."
var string t25 = "ATR multiplier for SuperTrend 4. Higher values place the band farther from price and reduce sensitivity."
var string t26 = "Smoothing method applied to the source before SuperTrend 4 is calculated."
var string t27 = "Length of the smoothing used for SuperTrend 4. Higher values smooth more but add lag."
var string t28 = "Relative influence of SuperTrend 4 inside the weighted cluster. Higher values make this member contribute more to the final consensus."
var string t29 = "ATR length for SuperTrend 5. Lower values react faster to price changes, while higher values make this member slower and smoother."
var string t30 = "ATR multiplier for SuperTrend 5. Higher values place the band farther from price and reduce sensitivity."
var string t31 = "Smoothing method applied to the source before SuperTrend 5 is calculated."
var string t32 = "Length of the smoothing used for SuperTrend 5. Higher values smooth more but add lag."
var string t33 = "Relative influence of SuperTrend 5 inside the weighted cluster. Higher values make this member contribute more to the final consensus."
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}
// ~~ INPUT PARAMETERS {
gCe = 'Cluster Engine'
thr = input.float(0.60, 'Consensus Threshold', minval = 0.0, maxval = 1.0, step = 0.01, group = gCe, tooltip = t1)
baseIx = input.int(3, 'Base SuperTrend Index', minval = 1, maxval = 5, group = gCe, tooltip = t2)
gVi = 'Visual Analytics'
useBc = input.bool(true, 'Dynamic Bar Coloring', group = gVi, tooltip = t3)
showLbl = input.bool(true, 'Show Cluster Labels', group = gVi, tooltip = t4)
showDot = input.bool(true, 'Show Base SuperTrend Flip Dots', group = gVi, tooltip = t5)
cBu = input.color(color.new(color.teal, 0), 'Bull', group = gVi, inline = "col", tooltip = t6)
cBe = input.color(color.new(#f7525f, 0), 'Bear', group = gVi, inline = "col", tooltip = t7)
cN = input.color(color.new(#ff9800, 20), 'Neutral', group = gVi, inline = "col", tooltip = t6+ "\n\n" +t7+ "\n\n" +t8)
// ~~ SuperTrend 1 {
gSt1 = 'SuperTrend 1'
a1 = input.int(7, 'ATR Length', minval = 1, group = gSt1, inline = '1', tooltip = t9)
f1 = input.float(1.5, 'Factor', minval = 0.01, step = 0.01, group = gSt1, inline = '1', tooltip = t9+ "\n\n" +t10)
m1 = input.string('EMA', 'Smoothing', options = ['SMA', 'EMA', 'DEMA', 'TEMA', 'LSMA', 'WMA', 'HMA', 'RMA'], group = gSt1, inline = '1.', tooltip = t11)
l1 = input.int(3, 'Length', minval = 1, group = gSt1, inline = '1.', tooltip = t11+ "\n\n" +t12)
w1 = input.float(1.0, 'Weight', minval = 0.0, step = 0.1, group = gSt1, inline = 'w1', tooltip = t13)
//}
// ~~ SuperTrend 2 {
gSt2 = 'SuperTrend 2'
a2 = input.int(10, 'ATR Length', minval = 1, group = gSt2, inline = '2', tooltip = t14)
f2 = input.float(2.0, 'Factor', minval = 0.01, step = 0.01, group = gSt2, inline = '2', tooltip = t14+ "\n\n" +t15)
m2 = input.string('EMA', 'Smoothing', options = ['SMA', 'EMA', 'DEMA', 'TEMA', 'LSMA', 'WMA', 'HMA', 'RMA'], group = gSt2, inline = '2.', tooltip = t16)
l2 = input.int(5, 'Length', minval = 1, group = gSt2, inline = '2.', tooltip = t16+ "\n\n" +t17)
w2 = input.float(1.0, 'Weight', minval = 0.0, step = 0.1, group = gSt2, inline = 'w2', tooltip = t18)
//}
// ~~ SuperTrend 3 {
gSt3 = 'SuperTrend 3'
a3 = input.int(14, 'ATR Length', minval = 1, group = gSt3, inline = '3', tooltip = t19)
f3 = input.float(2.5, 'Factor', minval = 0.01, step = 0.01, group = gSt3, inline = '3', tooltip = t19+ "\n\n" +t20)
m3 = input.string('SMA', 'Smoothing', options = ['SMA', 'EMA', 'DEMA', 'TEMA', 'LSMA', 'WMA', 'HMA', 'RMA'], group = gSt3, inline = '3.', tooltip = t21)
l3 = input.int(8, 'Length', minval = 1, group = gSt3, inline = '3.', tooltip =t21+ "\n\n" + t22)
w3 = input.float(1.2, 'Weight', minval = 0.0, step = 0.1, group = gSt3, inline = 'w3', tooltip = t23)
//}
// ~~ SuperTrend 4 {
gSt4 = 'SuperTrend 4'
a4 = input.int(21, 'ATR Length', minval = 1, group = gSt4, inline = '4', tooltip = t24)
f4 = input.float(3.0, 'Factor', minval = 0.01, step = 0.01, group = gSt4, inline = '4', tooltip = t24+ "\n\n" +t25)
m4 = input.string('WMA', 'Smoothing', options = ['SMA', 'EMA', 'DEMA', 'TEMA', 'LSMA', 'WMA', 'HMA', 'RMA'], group = gSt4, inline = '4.', tooltip = t26)
l4 = input.int(13, 'Length', minval = 1, group = gSt4, inline = '4.', tooltip = t26+ "\n\n" +t27)
w4 = input.float(1.4, 'Weight', minval = 0.0, step = 0.1, group = gSt4, inline = 'w4', tooltip = t28)
//}
// ~~ SuperTrend 5 {
gSt5 = 'SuperTrend 5'
a5 = input.int(34, 'ATR Length', minval = 1, group = gSt5, inline = '5', tooltip = t29)
f5 = input.float(4.0, 'Factor', minval = 0.01, step = 0.01, group = gSt5, inline = '5', tooltip = t29+ "\n\n" +t30)
m5 = input.string('HMA', 'Smoothing', options = ['SMA', 'EMA', 'DEMA', 'TEMA', 'LSMA', 'WMA', 'HMA', 'RMA'], group = gSt5, inline = '5.', tooltip = t31)
l5 = input.int(21, 'Length', minval = 1, group = gSt5, inline = '5.', tooltip = t31+ "\n\n" +t32)
w5 = input.float(1.6, 'Weight', minval = 0.0, step = 0.1, group = gSt5, inline = 'w5', tooltip = t33)
//}
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}
// ~~ CONSTANTS & STYLING {
EPS = 0.0000001
N = 5
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}
// ~~ HELPER FUNCTIONS {
fMa(t, s, l) =>
ln = math.max(1, l)
switch t
'SMA' => ta.sma(s, ln)
'EMA' => ta.ema(s, ln)
'LSMA' => ta.linreg(s, ln, 0)
'WMA' => ta.wma(s, ln)
'HMA' => ta.hma(s, ln)
'RMA' => ta.rma(s, ln)
=> ta.sma(s, ln)
fSt(src, atrLen, fac) =>
atr = ta.atr(math.max(1, atrLen))
ub0 = src + fac * atr
lb0 = src - fac * atr
ub = ub0
ub := na(ub[1]) ? ub0 : (ub0 < ub[1] or src[1] > ub[1] ? ub0 : ub[1])
lb = lb0
lb := na(lb[1]) ? lb0 : (lb0 > lb[1] or src[1] < lb[1] ? lb0 : lb[1])
d = 1.0
d := na(d[1]) ? 1.0 : d[1] == -1.0 and src > ub[1] ? 1.0 : d[1] == 1.0 and src < lb[1] ? -1.0 : d[1]
st = d == 1.0 ? lb : ub
[st, d]
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}
// ~~ MULTI-SUPERTREND ENGINE {
src = hlc3
s1 = fMa(m1, src, l1)
s2 = fMa(m2, src, l2)
s3 = fMa(m3, src, l3)
s4 = fMa(m4, src, l4)
s5 = fMa(m5, src, l5)
[st1, d1] = fSt(s1, a1, f1)
[st2, d2] = fSt(s2, a2, f2)
[st3, d3] = fSt(s3, a3, f3)
[st4, d4] = fSt(s4, a4, f4)
[st5, d5] = fSt(s5, a5, f5)
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}
// ~~ ARRAYS FOR STORAGE {
var array<float> wArr = array.new_float(0)
var array<float> stArr = array.new_float(0)
var array<float> dArr = array.new_float(0)
if barstate.isfirst
array.push(wArr, w1), array.push(wArr, w2), array.push(wArr, w3), array.push(wArr, w4), array.push(wArr, w5)
for _ = 0 to N - 1
array.push(stArr, na)
array.push(dArr, na)
if array.size(wArr) != N or array.size(stArr) != N or array.size(dArr) != N
runtime.error('Array size mismatch. Expected 5 elements in all arrays.')
array.set(stArr, 0, st1), array.set(stArr, 1, st2), array.set(stArr, 2, st3), array.set(stArr, 3, st4), array.set(stArr, 4, st5)
array.set(dArr, 0, d1), array.set(dArr, 1, d2), array.set(dArr, 2, d3), array.set(dArr, 3, d4), array.set(dArr, 4, d5)
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}
// ~~ CONSENSUS ENGINE {
var matrix<float> mDat = matrix.new<float>(N, 3, na)
if matrix.rows(mDat) != N or matrix.columns(mDat) != 3
runtime.error('Matrix size mismatch. Expected 5x3.')
for i = 0 to N - 1
matrix.set(mDat, i, 0, array.get(dArr, i))
matrix.set(mDat, i, 1, array.get(wArr, i))
matrix.set(mDat, i, 2, array.get(stArr, i))
wSum = 0.0
wBu = 0.0
wBe = 0.0
lnBuNum = 0.0
lnBeNum = 0.0
for i = 0 to N - 1
d = matrix.get(mDat, i, 0)
w = matrix.get(mDat, i, 1)
st = matrix.get(mDat, i, 2)
wSum += w
if d > 0
wBu += w
lnBuNum += st * w
else if d < 0
wBe += w
lnBeNum += st * w
wSum := math.max(wSum, EPS)
scBu = wBu / wSum
scBe = wBe / wSum
scCl = scBu - scBe
strCl = math.abs(scCl)
lnBu = wBu > 0 ? lnBuNum / wBu : na
lnBe = wBe > 0 ? lnBeNum / wBe : na
baseRow = math.max(0, math.min(N - 1, baseIx - 1))
stB = matrix.get(mDat, baseRow, 2)
dB = matrix.get(mDat, baseRow, 0)
flipBu = ta.crossover(dB, 0)
flipBe = ta.crossunder(dB, 0)
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}
// ~~ FINAL FILTERED REGIME {
isBu = scBu >= thr
isBe = scBe >= thr
okBu = (isBu and dB > 0)
okBe = (isBe and dB < 0)
var float dLast = 0.0
if okBu and not okBe
dLast := 1.0
else if okBe and not okBu
dLast := -1.0
lnCl = dLast > 0 ? lnBu : dLast < 0 ? lnBe : na
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}
// ~~ VISUALIZATION {
cBar = scCl > 0 ? color.from_gradient(strCl, 0.0, 1.0, cN, cBu) : color.from_gradient(strCl, 0.0, 1.0, cN, cBe)
barcolor(useBc ? cBar : na)
plotcandle(useBc?open:na,useBc?high:na,useBc?low: na,useBc?close:na, color=useBc?cBar:na, bordercolor = useBc?cBar:na,wickcolor=useBc?cBar:na)
plotshape(ta.crossover(dLast, 0), 'Major Long', shape.labelup, location.belowbar, color.new(cBu, 30), size = size.tiny, text = '▲', textcolor = color.white)
plotshape(ta.crossunder(dLast, 0), 'Major Short', shape.labeldown, location.abovebar, color.new(cBe, 30), size = size.tiny, text = '▼', textcolor = color.white)
plotshape(showDot and flipBu ? stB : na, 'Base ST Long', shape.triangleup, location.absolute, dLast > 0 ? color.new(cBu, 40) : color.new(cBe, 40), size = size.tiny)
plotshape(showDot and flipBe ? stB : na, 'Base ST Short', shape.triangledown, location.absolute, dLast > 0 ? color.new(cBu, 40) : color.new(cBe, 40), size = size.tiny)
if showLbl and flipBu
label.new(bar_index, stB, text = 'Bull Cluster\n' + str.tostring(scBu * 100.0, '#.#') + '%', color = color.new(cBu, 90), textcolor = cBu, style = label.style_label_up, yloc = yloc.price, size = size.small)
if showLbl and flipBe
label.new(bar_index, stB, text = 'Bear Cluster\n' + str.tostring(scBe * 100.0, '#.#') + '%', color = color.new(cBe, 90), textcolor = cBe, style = label.style_label_down, yloc = yloc.price, size = size.small)
pUp = plot(dLast == 1 ? lnCl : na, 'Cluster Up Trend', color = cBu, style = plot.style_linebr, linewidth = 2)
pDn = plot(dLast == -1 ? lnCl : na, 'Cluster Down Trend', color = cBe, style = plot.style_linebr, linewidth = 2)
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}
// ~~ ALERTS {
alBu = ta.crossover(dLast, 0)
alBe = ta.crossunder(dLast, 0)
alAny = alBu or alBe
alertcondition(alBu, 'Long', 'Bullish clustered SuperTrend signal')
alertcondition(alBe, 'Short', 'Bearish clustered SuperTrend signal')
alertcondition(alAny, 'Signal', 'Clustered SuperTrend signal')
//~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~}