mirror of
https://github.com/mihakralj/QuanTAlib.git
synced 2026-08-26 14:28:05 +00:00
Add TRAMA implementation and comprehensive tests
- Implemented the TRAMA (Trend Regularity Adaptive Moving Average) class with adaptive EMA logic. - Added unit tests for TRAMA functionality, including constructor validation, basic calculations, state management, and robustness checks. - Created validation tests to ensure consistency across different modes of operation (streaming, batch, and static calculations). - Enhanced documentation for TRAMA, including performance profiles and quality metrics. - Updated workspace configuration by removing unnecessary folder references.
This commit is contained in:
Miha Kralj
@@ -0,0 +1,254 @@
|
||||
namespace QuanTAlib.Tests;
|
||||
|
||||
public class SwmaValidationTests
|
||||
{
|
||||
private static TSeries MakeSeries(int count = 500)
|
||||
{
|
||||
var gbm = new GBM(startPrice: 100, seed: 42);
|
||||
var series = new TSeries();
|
||||
for (int i = 0; i < count; i++)
|
||||
{
|
||||
series.Add(gbm.Next());
|
||||
}
|
||||
return series;
|
||||
}
|
||||
|
||||
// === Self-consistency: Batch vs Streaming vs Span ===
|
||||
|
||||
[Fact]
|
||||
public void Batch_Matches_Streaming()
|
||||
{
|
||||
var src = MakeSeries(500);
|
||||
int period = 6;
|
||||
|
||||
var batchResult = Swma.Batch(src, period);
|
||||
|
||||
var streaming = new Swma(period);
|
||||
var streamResults = new TSeries();
|
||||
for (int i = 0; i < src.Count; i++)
|
||||
{
|
||||
streamResults.Add(streaming.Update(src[i]));
|
||||
}
|
||||
|
||||
for (int i = 0; i < src.Count; i++)
|
||||
{
|
||||
Assert.Equal(batchResult[i].Value, streamResults[i].Value, 1e-10);
|
||||
}
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public void Span_Matches_Streaming()
|
||||
{
|
||||
var src = MakeSeries(500);
|
||||
int period = 8;
|
||||
|
||||
var streaming = new Swma(period);
|
||||
var streamResults = new List<double>();
|
||||
for (int i = 0; i < src.Count; i++)
|
||||
{
|
||||
streamResults.Add(streaming.Update(src[i]).Value);
|
||||
}
|
||||
|
||||
var spanOutput = new double[src.Count];
|
||||
Swma.Batch(src.Values, spanOutput, period);
|
||||
|
||||
for (int i = 0; i < src.Count; i++)
|
||||
{
|
||||
Assert.Equal(streamResults[i], spanOutput[i], 1e-10);
|
||||
}
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public void Calculate_Matches_Batch()
|
||||
{
|
||||
var src = MakeSeries(300);
|
||||
int period = 5;
|
||||
|
||||
var batchResult = Swma.Batch(src, period);
|
||||
var (calcResult, _) = Swma.Calculate(src, period);
|
||||
|
||||
Assert.Equal(batchResult.Count, calcResult.Count);
|
||||
for (int i = 0; i < batchResult.Count; i++)
|
||||
{
|
||||
Assert.Equal(batchResult[i].Value, calcResult[i].Value, 1e-10);
|
||||
}
|
||||
}
|
||||
|
||||
// === Mathematical properties ===
|
||||
|
||||
[Fact]
|
||||
public void ConstantInput_ReturnsConstant_AllPeriods()
|
||||
{
|
||||
double constant = 42.0;
|
||||
int[] periods = { 2, 3, 4, 5, 10, 20 };
|
||||
|
||||
foreach (int period in periods)
|
||||
{
|
||||
var swma = new Swma(period);
|
||||
for (int i = 0; i < period + 5; i++)
|
||||
{
|
||||
swma.Update(new TValue(DateTime.UtcNow.AddSeconds(i), constant));
|
||||
}
|
||||
Assert.Equal(constant, swma.Last.Value, 1e-10);
|
||||
}
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public void OutputBounded_ByInputRange()
|
||||
{
|
||||
var src = MakeSeries(500);
|
||||
int period = 10;
|
||||
|
||||
var result = Swma.Batch(src, period);
|
||||
|
||||
// After warmup, output should be bounded by local window min/max
|
||||
for (int i = period - 1; i < src.Count; i++)
|
||||
{
|
||||
double min = double.MaxValue;
|
||||
double max = double.MinValue;
|
||||
for (int j = i - period + 1; j <= i; j++)
|
||||
{
|
||||
double v = src[j].Value;
|
||||
if (v < min) { min = v; }
|
||||
if (v > max) { max = v; }
|
||||
}
|
||||
Assert.InRange(result[i].Value, min - 1e-10, max + 1e-10);
|
||||
}
|
||||
}
|
||||
|
||||
[Theory]
|
||||
[InlineData(3)]
|
||||
[InlineData(5)]
|
||||
[InlineData(7)]
|
||||
[InlineData(11)]
|
||||
public void SymmetricWeights_SymmetricInput_ProducesCenter(int period)
|
||||
{
|
||||
// For symmetric weights and linearly increasing input fully filling the window,
|
||||
// the weighted average equals the center value
|
||||
var swma = new Swma(period);
|
||||
|
||||
for (int i = 0; i < period; i++)
|
||||
{
|
||||
swma.Update(new TValue(DateTime.UtcNow.AddSeconds(i), (double)(i + 1)));
|
||||
}
|
||||
|
||||
// Linear input [1..period]: center = (period+1)/2.0
|
||||
double expectedCenter = (period + 1) / 2.0;
|
||||
Assert.Equal(expectedCenter, swma.Last.Value, 1e-10);
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public void Period4_PineScript_Equivalence()
|
||||
{
|
||||
// PineScript ta.swma: weights [1, 2, 2, 1] / 6
|
||||
var swma = new Swma(period: 4);
|
||||
double[] values = { 100, 102, 98, 104, 106, 103, 101, 105 };
|
||||
var results = new List<double>();
|
||||
|
||||
for (int i = 0; i < values.Length; i++)
|
||||
{
|
||||
results.Add(swma.Update(new TValue(DateTime.UtcNow.AddSeconds(i), values[i])).Value);
|
||||
}
|
||||
|
||||
// Manual Pine calculation for bar 3 (index 3): (1*100 + 2*102 + 2*98 + 1*104)/6
|
||||
double expected3 = (100.0 + 204.0 + 196.0 + 104.0) / 6.0;
|
||||
Assert.Equal(expected3, results[3], 1e-10);
|
||||
|
||||
// bar 4: (1*102 + 2*98 + 2*104 + 1*106)/6
|
||||
double expected4 = (102.0 + 196.0 + 208.0 + 106.0) / 6.0;
|
||||
Assert.Equal(expected4, results[4], 1e-10);
|
||||
}
|
||||
|
||||
// === Stress and edge cases ===
|
||||
|
||||
[Fact]
|
||||
public void LargePeriod_Handles()
|
||||
{
|
||||
int period = 200;
|
||||
var src = MakeSeries(500);
|
||||
var result = Swma.Batch(src, period);
|
||||
|
||||
Assert.Equal(500, result.Count);
|
||||
Assert.True(double.IsFinite(result[^1].Value));
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public void AllNaN_Input_ReturnsNaN()
|
||||
{
|
||||
double[] source = new double[10];
|
||||
Array.Fill(source, double.NaN);
|
||||
double[] output = new double[10];
|
||||
|
||||
Swma.Batch(source, output, period: 3);
|
||||
|
||||
for (int i = 0; i < output.Length; i++)
|
||||
{
|
||||
Assert.True(double.IsNaN(output[i]));
|
||||
}
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public void MixedNaN_Recovers()
|
||||
{
|
||||
var swma = new Swma(period: 3);
|
||||
swma.Update(new TValue(DateTime.UtcNow, 10.0));
|
||||
swma.Update(new TValue(DateTime.UtcNow.AddSeconds(1), 20.0));
|
||||
swma.Update(new TValue(DateTime.UtcNow.AddSeconds(2), 30.0));
|
||||
|
||||
// Now NaN
|
||||
swma.Update(new TValue(DateTime.UtcNow.AddSeconds(3), double.NaN));
|
||||
Assert.True(double.IsFinite(swma.Last.Value));
|
||||
|
||||
// Recover with valid value
|
||||
swma.Update(new TValue(DateTime.UtcNow.AddSeconds(4), 40.0));
|
||||
Assert.True(double.IsFinite(swma.Last.Value));
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public void DifferentPeriods_ProduceDifferentResults()
|
||||
{
|
||||
var src = MakeSeries(100);
|
||||
|
||||
var r4 = Swma.Batch(src, 4);
|
||||
var r8 = Swma.Batch(src, 8);
|
||||
|
||||
// After both are hot, results should differ
|
||||
bool anyDifferent = false;
|
||||
for (int i = 20; i < src.Count; i++)
|
||||
{
|
||||
if (Math.Abs(r4[i].Value - r8[i].Value) > 1e-6)
|
||||
{
|
||||
anyDifferent = true;
|
||||
break;
|
||||
}
|
||||
}
|
||||
Assert.True(anyDifferent);
|
||||
}
|
||||
|
||||
[Fact]
|
||||
public void BarCorrection_ProducesSameAsNewSequence()
|
||||
{
|
||||
var src = MakeSeries(50);
|
||||
int period = 5;
|
||||
|
||||
// Path 1: All new bars
|
||||
var swma1 = new Swma(period);
|
||||
for (int i = 0; i < src.Count; i++)
|
||||
{
|
||||
swma1.Update(src[i], isNew: true);
|
||||
}
|
||||
|
||||
// Path 2: Bar correction on last bar
|
||||
var swma2 = new Swma(period);
|
||||
for (int i = 0; i < src.Count - 1; i++)
|
||||
{
|
||||
swma2.Update(src[i], isNew: true);
|
||||
}
|
||||
// Simulate tick corrections then final new bar
|
||||
swma2.Update(new TValue(DateTime.UtcNow, 999.0), isNew: true);
|
||||
swma2.Update(src[^1], isNew: false); // Correct last
|
||||
|
||||
// The correction path rewrites the last value
|
||||
Assert.Equal(swma1.Last.Value, swma2.Last.Value, 1e-10);
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user