Files
QuanTAlib/lib/numerics/betadist/tests/Betadist.Validation.Tests.cs
Miha Kralj 060649192f docs: remove C# Implementation Considerations sections, clean up temp scripts, reorganize test files
- Remove 'C# Implementation Considerations' sections from 34 indicator .md files
- Delete 29 temp PowerShell scripts (_fix_mojibake.ps1, _hex_scan.ps1, etc.)
- Move test files into tests/ subdirectories for consistent project structure
- Add trader-focused bullet points to indicator documentation
2026-03-12 12:34:16 -07:00

246 lines
9.3 KiB
C#

using Xunit;
namespace QuanTAlib.Tests;
/// <summary>
/// BetaDist validation tests — validates against known mathematical properties
/// of the regularized incomplete beta function. Known-value tests call
/// Betadist.IncompleteBeta directly (bypassing windowing) so results are exact.
/// Streaming/batch tests use GBM data and check invariants (bounds, finiteness,
/// monotonicity, symmetry) that hold regardless of window state.
/// </summary>
public class BetadistValidationTests
{
private const double Tolerance = 1e-9;
private const double LooseTolerance = 1e-6;
// ─── Mathematical invariants (invariant to normalization) ────────────────
[Fact]
public void BetaCdf_FlatRange_ReturnsHalf()
{
// When all window values are equal → range=0 → x=0.5
// For symmetric distributions (alpha=beta), CDF(0.5) = 0.5
double[] shapes = { 0.5, 1.0, 2.0, 3.0, 5.0 };
var time = DateTime.UtcNow;
foreach (double shape in shapes)
{
var ind = new Betadist(20, shape, shape);
for (int i = 0; i < 20; i++)
{
ind.Update(new TValue(time.AddSeconds(i), 100.0));
}
Assert.Equal(0.5, ind.Last.Value, LooseTolerance);
}
}
[Fact]
public void BetaCdf_OutputBounded_Zero_To_One()
{
int count = 200;
var gbm = new GBM(startPrice: 100, mu: 0.05, sigma: 0.3, seed: 51001);
var bars = gbm.Fetch(count, DateTime.UtcNow.Ticks, TimeSpan.FromMinutes(1));
var indicator = new Betadist(period: 20, alpha: 2.0, beta: 2.0);
for (int i = 0; i < count; i++)
{
indicator.Update(bars.Close[i]);
double v = indicator.Last.Value;
Assert.True(v >= 0.0 && v <= 1.0, $"Output {v} at bar {i} out of [0,1]");
}
}
// ─── Period=1 trivial case ────────────────────────────────────────────────
[Fact]
public void BetaCdf_Period1_AlwaysReturnsCdfAtHalf()
{
// period=1: single-element window → range=0 → x=0.5 always
// CDF(0.5, 1, 1) = 0.5 exactly (uniform)
var ind = new Betadist(1, 1.0, 1.0);
var time = DateTime.UtcNow;
double[] prices = { 100.0, 50.0, 200.0, 1.0, 1000.0 };
foreach (double p in prices)
{
ind.Update(new TValue(time, p));
time = time.AddMinutes(1);
Assert.Equal(0.5, ind.Last.Value, LooseTolerance);
}
}
// ─── Known-value tests via IncompleteBeta static method ──────────────────
// These bypass windowing entirely and test the CDF math directly.
[Theory]
[InlineData(0.25, 2.0, 2.0, 0.15625)] // Beta(2,2): I(0.25) = 0.15625
[InlineData(0.5, 2.0, 2.0, 0.5)] // Beta(2,2): I(0.5) = 0.5 (symmetry)
[InlineData(0.75, 2.0, 2.0, 0.84375)] // Beta(2,2): I(0.75) = 0.84375
[InlineData(0.5, 2.0, 3.0, 0.6875)] // Beta(2,3): I(0.5) = 0.6875
[InlineData(0.5, 3.0, 2.0, 0.3125)] // Beta(3,2): I(0.5) = 0.3125
[InlineData(0.5, 1.0, 1.0, 0.5)] // Uniform: I(0.5) = 0.5
[InlineData(0.25, 1.0, 1.0, 0.25)] // Uniform: I(0.25) = 0.25
[InlineData(0.75, 1.0, 1.0, 0.75)] // Uniform: I(0.75) = 0.75
public void BetaCdf_IncompleteBeta_KnownValues(double x, double alpha, double beta, double expected)
{
double actual = Betadist.IncompleteBeta(x, alpha, beta);
Assert.Equal(expected, actual, LooseTolerance);
}
// ─── Complementary symmetry: I_x(a,b) + I_{1-x}(b,a) = 1 ───────────────
[Theory]
[InlineData(0.3, 2.0, 3.0)]
[InlineData(0.7, 2.0, 3.0)]
[InlineData(0.5, 1.5, 4.0)]
[InlineData(0.2, 3.0, 5.0)]
public void BetaCdf_ComplementarySymmetry(double x, double alpha, double beta)
{
double iab = Betadist.IncompleteBeta(x, alpha, beta);
double iba = Betadist.IncompleteBeta(1.0 - x, beta, alpha);
Assert.Equal(1.0, iab + iba, LooseTolerance);
}
// ─── Monotonicity via direct CDF ─────────────────────────────────────────
[Fact]
public void BetaCdf_MonotonicIncreasing()
{
// CDF must be non-decreasing as x increases from 0 to 1
double alpha = 2.0, beta = 2.0;
double prevCdf = -1.0;
for (int i = 0; i <= 10; i++)
{
double x = i / 10.0 + 1e-10; // avoid exact 0
x = Math.Min(x, 1.0 - 1e-10);
double cdf = Betadist.IncompleteBeta(x, alpha, beta);
Assert.True(cdf >= prevCdf - LooseTolerance,
$"CDF not monotonic at x={x}: got {cdf}, prev={prevCdf}");
prevCdf = cdf;
}
}
// ─── Span batch consistency ───────────────────────────────────────────────
[Fact]
public void Batch_Span_MatchesTSeries()
{
int count = 150;
var gbm = new GBM(startPrice: 100, mu: 0.05, sigma: 0.25, seed: 51002);
var bars = gbm.Fetch(count, DateTime.UtcNow.Ticks, TimeSpan.FromMinutes(1));
double[] rawValues = new double[count];
for (int i = 0; i < count; i++)
{
rawValues[i] = bars.Close[i].Value;
}
var tseriesResult = Betadist.Batch(bars.Close, period: 30);
double[] spanResult = new double[count];
Betadist.Batch(rawValues, spanResult, period: 30);
for (int i = 0; i < count; i++)
{
Assert.Equal(tseriesResult[i].Value, spanResult[i], Tolerance);
}
}
// ─── Streaming convergence ────────────────────────────────────────────────
[Fact]
public void BetaCdf_HighPeriod_StillConverges()
{
int period = 200;
var indicator = new Betadist(period, 2.0, 5.0);
var gbm = new GBM(startPrice: 100, mu: 0.05, sigma: 0.3, seed: 51003);
var bars = gbm.Fetch(period + 50, DateTime.UtcNow.Ticks, TimeSpan.FromMinutes(1));
for (int i = 0; i < bars.Close.Count; i++)
{
indicator.Update(bars.Close[i]);
Assert.True(double.IsFinite(indicator.Last.Value),
$"Non-finite output at bar {i}");
}
}
[Fact]
public void BetaCdf_ExtremePrices_StillInRange()
{
var indicator = new Betadist(period: 20, alpha: 2.0, beta: 2.0);
var time = DateTime.UtcNow;
for (int i = 0; i < 20; i++)
{
double price = (i % 2 == 0) ? 1e10 : 1e-10;
indicator.Update(new TValue(time.AddMinutes(i), price));
double v = indicator.Last.Value;
Assert.True(v >= 0.0 && v <= 1.0, $"Out of range at {i}: {v}");
}
}
// ─── Different parameter combos all produce output in range ──────────────
[Theory]
[InlineData(5, 0.5, 0.5)]
[InlineData(14, 1.0, 1.0)]
[InlineData(50, 2.0, 2.0)]
[InlineData(100, 3.0, 5.0)]
[InlineData(30, 0.5, 2.0)]
public void BetaCdf_ParameterCombos_OutputBounded(int period, double alpha, double beta)
{
int count = period + 50;
var gbm = new GBM(startPrice: 100, mu: 0.05, sigma: 0.2, seed: 51004 + period);
var bars = gbm.Fetch(count, DateTime.UtcNow.Ticks, TimeSpan.FromMinutes(1));
var indicator = new Betadist(period, alpha, beta);
for (int i = 0; i < count; i++)
{
indicator.Update(bars.Close[i]);
double v = indicator.Last.Value;
Assert.True(v >= 0.0 && v <= 1.0);
}
}
// ─── Large dataset: stable ────────────────────────────────────────────────
[Fact]
public void BetaCdf_LargeDataset_Stable()
{
int count = 2000;
var gbm = new GBM(startPrice: 100, mu: 0.05, sigma: 0.2, seed: 51005);
var bars = gbm.Fetch(count, DateTime.UtcNow.Ticks, TimeSpan.FromMinutes(1));
var indicator = new Betadist(period: 50);
for (int i = 0; i < count; i++)
{
indicator.Update(bars.Close[i]);
double v = indicator.Last.Value;
Assert.True(double.IsFinite(v) && v >= 0.0 && v <= 1.0,
$"Invalid output {v} at bar {i}");
}
}
// ─── Alpha != Beta produces asymmetric CDF ───────────────────────────────
[Fact]
public void BetaCdf_AsymmetricParams_SkewsOutput()
{
// Beta(0.5, 5): mode near 0, most mass below 0.5 → CDF(0.5) > 0.5
// Beta(5, 0.5): mode near 1, most mass above 0.5 → CDF(0.5) < 0.5
double cdfLow = Betadist.IncompleteBeta(0.5, 0.5, 5.0);
double cdfHigh = Betadist.IncompleteBeta(0.5, 5.0, 0.5);
Assert.True(cdfLow > cdfHigh,
$"Beta(0.5,5) CDF at 0.5 ({cdfLow:F6}) should be > Beta(5,0.5) ({cdfHigh:F6})");
Assert.True(cdfLow > 0.5, $"Beta(0.5,5) CDF(0.5)={cdfLow} should be > 0.5");
Assert.True(cdfHigh < 0.5, $"Beta(5,0.5) CDF(0.5)={cdfHigh} should be < 0.5");
}
}