mirror of
https://github.com/mihakralj/QuanTAlib.git
synced 2026-07-27 17:27:43 +00:00
67ad6f0cba
Comprehensive refactor across all indicators replacing the periodic ResyncInterval-based drift correction (every 1000 ticks recalculate from scratch) with Kahan compensated summation for running sums. Key changes: - Remove ResyncInterval constants and TickCount fields from all State records - Add Kahan compensation fields (SumComp, SumSqComp, etc.) to State records - Replace naive sum += val - removed with Kahan delta pattern - Remove Resync()/RecalculateSum() methods that did O(N) recalculation - Update batch/SIMD paths to use Kahan compensation instead of resync loops - IIR filters (EMA, REMA, RGMA) simplified: inherently self-correcting - Version bump to 0.8.7 - Build system: README version stamping via Directory.Build.props - Minor doc/test tolerance adjustments for new numerical characteristics Affected modules: channels, core, cycles, dynamics, errors, momentum, oscillators, statistics, trends_FIR, trends_IIR, volatility, volume
414 lines
13 KiB
C#
414 lines
13 KiB
C#
using System.Runtime.CompilerServices;
|
||
using System.Runtime.InteropServices;
|
||
|
||
namespace QuanTAlib;
|
||
|
||
/// <summary>
|
||
/// BBWP: Bollinger Band Width Percentile
|
||
/// </summary>
|
||
/// <remarks>
|
||
/// BBWP measures where the current Bollinger Band Width falls within its
|
||
/// historical distribution, expressing the result as a percentile rank
|
||
/// between 0 and 1. Unlike BBWN which normalizes using min/max values,
|
||
/// BBWP uses percentile ranking which is more robust to outliers.
|
||
///
|
||
/// Formula:
|
||
/// <c>BBW = 2 × multiplier × StdDev(source, period)</c>
|
||
/// <c>BBWP = count(BBW_history < BBW_current) / total_count</c>
|
||
///
|
||
/// The indicator first calculates the standard BBW, then determines what
|
||
/// percentage of historical BBW values fall below the current value.
|
||
/// Values near 0 indicate current volatility is lower than most historical
|
||
/// readings, while values near 1 indicate it's higher than most.
|
||
///
|
||
/// Key properties:
|
||
/// - Range: [0, 1] (percentile)
|
||
/// - 0.0 indicates current BBW is lowest in lookback period
|
||
/// - 1.0 indicates current BBW is highest in lookback period
|
||
/// - 0.5 indicates median volatility when no percentile can be calculated
|
||
/// </remarks>
|
||
[SkipLocalsInit]
|
||
public sealed class Bbwp : AbstractBase
|
||
{
|
||
private readonly int _period;
|
||
private readonly double _multiplier;
|
||
private readonly int _lookback;
|
||
private readonly RingBuffer _buffer;
|
||
private readonly RingBuffer _bbwBuffer;
|
||
|
||
[StructLayout(LayoutKind.Auto)]
|
||
private record struct State(
|
||
double Sum,
|
||
double SumSq,
|
||
double SumComp,
|
||
double SumSqComp,
|
||
double LastValid);
|
||
private State _state;
|
||
private State _p_state;
|
||
|
||
/// <summary>
|
||
/// Creates BBWP with specified period, multiplier, and lookback.
|
||
/// </summary>
|
||
/// <param name="period">Lookback period for BB calculations (must be > 0)</param>
|
||
/// <param name="multiplier">Standard deviation multiplier (must be > 0)</param>
|
||
/// <param name="lookback">Historical lookback period for percentile calculation (must be > 0)</param>
|
||
public Bbwp(int period, double multiplier = 2.0, int lookback = 252)
|
||
{
|
||
if (period <= 0)
|
||
{
|
||
throw new ArgumentException("Period must be greater than 0", nameof(period));
|
||
}
|
||
|
||
if (multiplier <= 0)
|
||
{
|
||
throw new ArgumentException("Multiplier must be greater than 0", nameof(multiplier));
|
||
}
|
||
|
||
if (lookback <= 0)
|
||
{
|
||
throw new ArgumentException("Lookback must be greater than 0", nameof(lookback));
|
||
}
|
||
|
||
_period = period;
|
||
_multiplier = multiplier;
|
||
_lookback = lookback;
|
||
_buffer = new RingBuffer(period);
|
||
_bbwBuffer = new RingBuffer(lookback);
|
||
Name = $"Bbwp({period},{multiplier:F1},{lookback})";
|
||
WarmupPeriod = period + lookback;
|
||
}
|
||
|
||
/// <summary>
|
||
/// Creates BBWP with specified source, period, multiplier, and lookback.
|
||
/// </summary>
|
||
public Bbwp(ITValuePublisher source, int period, double multiplier = 2.0, int lookback = 252) : this(period, multiplier, lookback)
|
||
{
|
||
source.Pub += Handle;
|
||
}
|
||
|
||
private void Handle(object? sender, in TValueEventArgs e) => Update(e.Value, e.IsNew);
|
||
|
||
/// <summary>
|
||
/// True if the indicator has enough data for valid results.
|
||
/// </summary>
|
||
public override bool IsHot => _buffer.IsFull && _bbwBuffer.Count >= Math.Min(10, _lookback);
|
||
|
||
/// <summary>
|
||
/// Period of the indicator.
|
||
/// </summary>
|
||
public int Period => _period;
|
||
|
||
/// <summary>
|
||
/// Standard deviation multiplier.
|
||
/// </summary>
|
||
public double Multiplier => _multiplier;
|
||
|
||
/// <summary>
|
||
/// Historical lookback period for percentile calculation.
|
||
/// </summary>
|
||
public int Lookback => _lookback;
|
||
[MethodImpl(MethodImplOptions.AggressiveInlining)]
|
||
public override TValue Update(TValue input, bool isNew = true)
|
||
{
|
||
double value = input.Value;
|
||
|
||
// Sanitize input
|
||
if (!double.IsFinite(value))
|
||
{
|
||
value = double.IsFinite(_state.LastValid) ? _state.LastValid : 0.0;
|
||
}
|
||
else
|
||
{
|
||
_state.LastValid = value;
|
||
}
|
||
|
||
if (isNew)
|
||
{
|
||
_p_state = _state;
|
||
|
||
// Kahan compensated sliding window update
|
||
if (_buffer.Count == _buffer.Capacity)
|
||
{
|
||
double oldest = _buffer.Oldest;
|
||
double delta = value - oldest;
|
||
{
|
||
double y = delta - _state.SumComp;
|
||
double t = _state.Sum + y;
|
||
_state.SumComp = (t - _state.Sum) - y;
|
||
_state.Sum = t;
|
||
}
|
||
{
|
||
double deltaSq = (value * value) - (oldest * oldest);
|
||
double y = deltaSq - _state.SumSqComp;
|
||
double t = _state.SumSq + y;
|
||
_state.SumSqComp = (t - _state.SumSq) - y;
|
||
_state.SumSq = t;
|
||
}
|
||
}
|
||
else
|
||
{
|
||
{
|
||
double y = value - _state.SumComp;
|
||
double t = _state.Sum + y;
|
||
_state.SumComp = (t - _state.Sum) - y;
|
||
_state.Sum = t;
|
||
}
|
||
{
|
||
double sq = value * value;
|
||
double y = sq - _state.SumSqComp;
|
||
double t = _state.SumSq + y;
|
||
_state.SumSqComp = (t - _state.SumSq) - y;
|
||
_state.SumSq = t;
|
||
}
|
||
}
|
||
|
||
_buffer.Add(value);
|
||
}
|
||
else
|
||
{
|
||
_state = _p_state;
|
||
|
||
// Update the newest value in buffer
|
||
_buffer.UpdateNewest(value);
|
||
RecalculateSums();
|
||
}
|
||
|
||
// Calculate BBW first
|
||
int count = _buffer.Count;
|
||
if (count == 0)
|
||
{
|
||
Last = new TValue(input.Time, 0.5);
|
||
PubEvent(Last, isNew);
|
||
return Last;
|
||
}
|
||
|
||
double mean = _state.Sum / count;
|
||
double variance = Math.Max(0.0, (_state.SumSq / count) - (mean * mean));
|
||
double stddev = Math.Sqrt(variance);
|
||
double bbw = 2.0 * _multiplier * stddev;
|
||
|
||
// Add BBW to history buffer for percentile calculation
|
||
if (isNew)
|
||
{
|
||
_bbwBuffer.Add(bbw);
|
||
}
|
||
else
|
||
{
|
||
_bbwBuffer.UpdateNewest(bbw);
|
||
}
|
||
|
||
// Calculate percentile of current BBW within historical distribution
|
||
double bbwp = 0.5; // Default when no percentile can be calculated
|
||
int totalCount = _bbwBuffer.Count;
|
||
if (totalCount >= 1)
|
||
{
|
||
int countBelow = 0;
|
||
for (int i = 0; i < totalCount; i++)
|
||
{
|
||
if (_bbwBuffer[i] < bbw)
|
||
{
|
||
countBelow++;
|
||
}
|
||
}
|
||
bbwp = (double)countBelow / totalCount;
|
||
}
|
||
|
||
// Clamp to [0,1] range (should already be in range, but ensure safety)
|
||
bbwp = Math.Max(0.0, Math.Min(1.0, bbwp));
|
||
|
||
Last = new TValue(input.Time, bbwp);
|
||
PubEvent(Last, isNew);
|
||
return Last;
|
||
}
|
||
public override TSeries Update(TSeries source)
|
||
{
|
||
int len = source.Count;
|
||
var t = new List<long>(len);
|
||
var v = new List<double>(len);
|
||
CollectionsMarshal.SetCount(t, len);
|
||
CollectionsMarshal.SetCount(v, len);
|
||
|
||
var tSpan = CollectionsMarshal.AsSpan(t);
|
||
var vSpan = CollectionsMarshal.AsSpan(v);
|
||
|
||
Batch(source.Values, vSpan, _period, _multiplier, _lookback);
|
||
source.Times.CopyTo(tSpan);
|
||
|
||
// Update internal state to match final position
|
||
for (int i = 0; i < len; i++)
|
||
{
|
||
Update(new TValue(source.Times[i], source.Values[i]), isNew: true);
|
||
}
|
||
|
||
return new TSeries(t, v);
|
||
}
|
||
|
||
[MethodImpl(MethodImplOptions.AggressiveInlining)]
|
||
private void RecalculateSums()
|
||
{
|
||
_state.Sum = 0.0;
|
||
_state.SumSq = 0.0;
|
||
for (int i = 0; i < _buffer.Count; i++)
|
||
{
|
||
double v = _buffer[i];
|
||
_state.Sum += v;
|
||
_state.SumSq += v * v;
|
||
}
|
||
}
|
||
public override void Prime(ReadOnlySpan<double> source, TimeSpan? step = null)
|
||
{
|
||
for (int i = 0; i < source.Length; i++)
|
||
{
|
||
Update(new TValue(DateTime.UtcNow, source[i]), isNew: true);
|
||
}
|
||
}
|
||
public override void Reset()
|
||
{
|
||
_buffer.Clear();
|
||
_bbwBuffer.Clear();
|
||
_state = default;
|
||
_p_state = default;
|
||
Last = default;
|
||
}
|
||
|
||
/// <summary>
|
||
/// Calculates BBWP for entire series.
|
||
/// </summary>
|
||
public static TSeries Batch(TSeries source, int period, double multiplier = 2.0, int lookback = 252)
|
||
{
|
||
if (period <= 0)
|
||
{
|
||
throw new ArgumentException("Period must be greater than 0", nameof(period));
|
||
}
|
||
|
||
if (multiplier <= 0)
|
||
{
|
||
throw new ArgumentException("Multiplier must be greater than 0", nameof(multiplier));
|
||
}
|
||
|
||
if (lookback <= 0)
|
||
{
|
||
throw new ArgumentException("Lookback must be greater than 0", nameof(lookback));
|
||
}
|
||
|
||
int len = source.Count;
|
||
var t = new List<long>(len);
|
||
var v = new List<double>(len);
|
||
CollectionsMarshal.SetCount(t, len);
|
||
CollectionsMarshal.SetCount(v, len);
|
||
|
||
var tSpan = CollectionsMarshal.AsSpan(t);
|
||
var vSpan = CollectionsMarshal.AsSpan(v);
|
||
|
||
Batch(source.Values, vSpan, period, multiplier, lookback);
|
||
source.Times.CopyTo(tSpan);
|
||
|
||
return new TSeries(t, v);
|
||
}
|
||
|
||
/// <summary>
|
||
/// Batch BBWP calculation with O(1) rolling variance and percentile ranking.
|
||
/// </summary>
|
||
[MethodImpl(MethodImplOptions.AggressiveInlining)]
|
||
public static void Batch(ReadOnlySpan<double> source, Span<double> output, int period, double multiplier = 2.0, int lookback = 252)
|
||
{
|
||
if (source.Length != output.Length)
|
||
{
|
||
throw new ArgumentException("Source and output must have the same length", nameof(output));
|
||
}
|
||
|
||
if (period <= 0)
|
||
{
|
||
throw new ArgumentException("Period must be greater than 0", nameof(period));
|
||
}
|
||
|
||
if (multiplier <= 0)
|
||
{
|
||
throw new ArgumentException("Multiplier must be greater than 0", nameof(multiplier));
|
||
}
|
||
|
||
if (lookback <= 0)
|
||
{
|
||
throw new ArgumentException("Lookback must be greater than 0", nameof(lookback));
|
||
}
|
||
|
||
int len = source.Length;
|
||
if (len == 0)
|
||
{
|
||
return;
|
||
}
|
||
|
||
double sum = 0.0;
|
||
double sumSq = 0.0;
|
||
double mult2 = 2.0 * multiplier;
|
||
double lastValid = 0.0;
|
||
var bbwHistory = new RingBuffer(lookback);
|
||
|
||
// Buffer to track sanitized values for correct window removal
|
||
var valueBuffer = new RingBuffer(period);
|
||
|
||
for (int i = 0; i < len; i++)
|
||
{
|
||
double val = source[i];
|
||
|
||
// Sanitize input - mirror Update method behavior
|
||
if (!double.IsFinite(val))
|
||
{
|
||
val = lastValid;
|
||
}
|
||
else
|
||
{
|
||
lastValid = val;
|
||
}
|
||
|
||
// Remove oldest sanitized value if past warmup
|
||
if (i >= period)
|
||
{
|
||
double oldest = valueBuffer.Oldest;
|
||
sum -= oldest;
|
||
sumSq -= oldest * oldest;
|
||
}
|
||
|
||
// Add new sanitized value
|
||
sum += val;
|
||
sumSq += val * val;
|
||
valueBuffer.Add(val);
|
||
|
||
// Calculate BBW
|
||
int count = Math.Min(i + 1, period);
|
||
double mean = sum / count;
|
||
double variance = Math.Max(0.0, (sumSq / count) - (mean * mean));
|
||
double stddev = Math.Sqrt(variance);
|
||
double bbw = mult2 * stddev;
|
||
|
||
// Add to BBW history
|
||
bbwHistory.Add(bbw);
|
||
|
||
// Calculate percentile of current BBW within historical distribution
|
||
double bbwp = 0.5; // Default
|
||
int totalCount = bbwHistory.Count;
|
||
if (totalCount >= 1)
|
||
{
|
||
int countBelow = 0;
|
||
for (int j = 0; j < totalCount; j++)
|
||
{
|
||
if (bbwHistory[j] < bbw)
|
||
{
|
||
countBelow++;
|
||
}
|
||
}
|
||
bbwp = (double)countBelow / totalCount;
|
||
}
|
||
|
||
// Clamp to [0,1] range
|
||
output[i] = Math.Max(0.0, Math.Min(1.0, bbwp));
|
||
}
|
||
}
|
||
|
||
public static (TSeries Results, Bbwp Indicator) Calculate(TSeries source, int period, double multiplier = 2.0, int lookback = 252)
|
||
{
|
||
var indicator = new Bbwp(period, multiplier, lookback);
|
||
TSeries results = indicator.Update(source);
|
||
return (results, indicator);
|
||
}
|
||
} |