Migrated StudentT distribution to shape fitter

FastRng/Double/Distributions/StudentT.cs

@@ -1,36 +0,0 @@
-using System;
-using System.Threading;
-using System.Threading.Tasks;
-
-namespace FastRng.Double.Distributions
-{
-    public sealed class StudentT : IDistribution
-    {
-        private static readonly IDistribution NORMAL_DISTRIBUTED = new Normal();
-        private double degreesOfFreedom = 1.0;
-        
-        public IRandom Random { get; set; }
-        
-        public double DegreesOfFreedom
-        {
-            get => this.degreesOfFreedom;
-            set
-            {
-                if(value <= 0.0)
-                    throw new ArgumentOutOfRangeException(message: "DegreesOfFreedom must be greater than 0", null);
-                
-                this.degreesOfFreedom = value;
-            }
-        }
-        
-        public async ValueTask<double> GetDistributedValue(CancellationToken token = default)
-        {
-            if (this.Random == null)
-                return double.NaN;
-
-            var normal = await this.Random.NextNumber(NORMAL_DISTRIBUTED, token);
-            var chiSquare = await this.Random.NextNumber(new ChiSquare {DegreesOfFreedom = this.DegreesOfFreedom}, token);
-            return normal / Math.Sqrt(chiSquare / this.DegreesOfFreedom);
-        }
-    }
-}

FastRng/Double/Distributions/StudentTNu1.cs

@@ -0,0 +1,48 @@
+using System;
+using System.Threading;
+using System.Threading.Tasks;
+
+namespace FastRng.Double.Distributions
+{
+    public sealed class StudentTNu1 : IDistribution
+    {
+        private const double NU = 1.0;
+        private const double START = 0.0;
+        private const double COMPRESS = 1.0;
+        private const double CONSTANT = 3.14190548592729;
+        
+        private static readonly double DIVIDEND;
+        private static readonly double DIVISOR;
+        private static readonly double EXPONENT;
+        
+        private ShapeFitter fitter;
+        private IRandom random;
+
+        static StudentTNu1()
+        {
+            DIVIDEND = MathTools.Gamma((NU + 1.0d) * 0.5d);
+            DIVISOR = Math.Sqrt(NU * Math.PI) * MathTools.Gamma(NU * 0.5d);
+            EXPONENT = -((NU + 1.0d) * 0.5d);
+        }
+        
+        public IRandom Random
+        {
+            get => this.random;
+            set
+            {
+                this.random = value;
+                this.fitter = new ShapeFitter(StudentTNu1.ShapeFunction, this.random, 100);
+            }
+        }
+
+        private static double ShapeFunction(double x) => CONSTANT * Math.Pow((DIVIDEND / DIVISOR) * Math.Pow(1.0d + Math.Pow(START + x * COMPRESS, 2) / NU, EXPONENT), COMPRESS);
+
+        public async ValueTask<double> GetDistributedValue(CancellationToken token = default)
+        {
+            if (this.Random == null)
+                return double.NaN;
+            
+            return await this.fitter.NextNumber(token);
+        }
+    }
+}

FastRngTests/Double/Distributions/StudentTNu1.cs

@@ -8,30 +8,40 @@ using NUnit.Framework;
 namespace FastRngTests.Double.Distributions
 {
     [ExcludeFromCodeCoverage]
-    public class StudentT
+    public class StudentTNu1
     {
         [Test]
         [Category(TestCategories.COVER)]
         [Category(TestCategories.NORMAL)]
         public async Task TestStudentTDistribution01()
         {
-            const double DOF = 6;
+            var dist = new FastRng.Double.Distributions.StudentTNu1();
-            const double MEAN = 0.0;
+            var fra = new FrequencyAnalysis();
-            const double VARIANCE = DOF / (DOF - 2);
-            
-            var dist = new FastRng.Double.Distributions.StudentT{ DegreesOfFreedom = DOF };
-            var stats = new RunningStatistics();
             var rng = new MultiThreadedRng();
             
             for (var n = 0; n < 100_000; n++)
-                stats.Push(await rng.NextNumber(dist));
+                fra.CountThis(await rng.NextNumber(dist));
             
             rng.StopProducer();
-            TestContext.WriteLine($"mean={MEAN} vs. {stats.Mean}");
+            var result = fra.NormalizeAndPlotEvents(TestContext.WriteLine);
-            TestContext.WriteLine($"variance={VARIANCE} vs {stats.Variance}");
+
+            Assert.That(result[0], Is.EqualTo(1.000000000).Within(0.2));
+            Assert.That(result[1], Is.EqualTo(0.999700120).Within(0.2));
+            Assert.That(result[2], Is.EqualTo(0.999200719).Within(0.2));
             
-            Assert.That(stats.Mean, Is.EqualTo(MEAN).Within(0.1), "Mean is out of range");
+            Assert.That(result[21], Is.EqualTo(0.953929798).Within(0.2));
-            Assert.That(stats.Variance, Is.EqualTo(VARIANCE).Within(0.1), "Variance is out of range");
+            Assert.That(result[22], Is.EqualTo(0.949852788).Within(0.2));
+            Assert.That(result[23], Is.EqualTo(0.945631619).Within(0.2));
+            
+            Assert.That(result[50], Is.EqualTo(0.793667169).Within(0.095));
+            
+            Assert.That(result[75], Is.EqualTo(0.633937627).Within(0.09));
+            Assert.That(result[85], Is.EqualTo(0.574902276).Within(0.09));
+            Assert.That(result[90], Is.EqualTo(0.547070729).Within(0.09));
+            
+            Assert.That(result[97], Is.EqualTo(0.510150990).Within(0.09));
+            Assert.That(result[98], Is.EqualTo(0.505075501).Within(0.09));
+            Assert.That(result[99], Is.EqualTo(0.500050000).Within(0.09));
         }
         
         [Test]
@@ -40,9 +50,10 @@ namespace FastRngTests.Double.Distributions
         public async Task TestStudentTGeneratorWithRange01()
         {
             var rng = new MultiThreadedRng();
+            var dist = new FastRng.Double.Distributions.StudentTNu1();
             var samples = new double[1_000];
             for (var n = 0; n < samples.Length; n++)
-                samples[n] = await rng.NextNumber(-1.0, 1.0, new FastRng.Double.Distributions.StudentT());
+                samples[n] = await rng.NextNumber(-1.0, 1.0, dist);
             
             rng.StopProducer();
             Assert.That(samples.Min(), Is.GreaterThanOrEqualTo(-1.0), "Min out of range");
@@ -55,9 +66,10 @@ namespace FastRngTests.Double.Distributions
         public async Task TestStudentTGeneratorWithRange02()
         {
             var rng = new MultiThreadedRng();
+            var dist = new FastRng.Double.Distributions.StudentTNu1();
             var samples = new double[1_000];
             for (var n = 0; n < samples.Length; n++)
-                samples[n] = await rng.NextNumber(0.0, 1.0, new FastRng.Double.Distributions.StudentT());
+                samples[n] = await rng.NextNumber(0.0, 1.0, dist);
             
             rng.StopProducer();
             Assert.That(samples.Min(), Is.GreaterThanOrEqualTo(0.0), "Min is out of range");
@@ -70,7 +82,7 @@ namespace FastRngTests.Double.Distributions
         public async Task TestStudentTGeneratorWithRange03()
         {
             var rng = new MultiThreadedRng();
-            var dist = new FastRng.Double.Distributions.StudentT { Random = rng }; // Test default parameters
+            var dist = new FastRng.Double.Distributions.StudentTNu1 { Random = rng }; // Test default parameters
             
             var samples = new double[1_000];
             for (var n = 0; n < samples.Length; n++)
@@ -81,25 +93,12 @@ namespace FastRngTests.Double.Distributions
             Assert.That(samples.Max(), Is.LessThanOrEqualTo(1.0), "Max is out of range");
         }
 
-        [Test]
-        [Category(TestCategories.COVER)]
-        [Category(TestCategories.NORMAL)]
-        public void ParameterTest01()
-        {
-            var dist = new FastRng.Double.Distributions.StudentT();
-            
-            Assert.Throws<ArgumentOutOfRangeException>(() => dist.DegreesOfFreedom = 0);
-            Assert.Throws<ArgumentOutOfRangeException>(() => dist.DegreesOfFreedom = -78);
-            Assert.DoesNotThrow(() => dist.DegreesOfFreedom = 0.0001);
-            Assert.DoesNotThrow(() => dist.DegreesOfFreedom = 4);
-        }
-
         [Test]
         [Category(TestCategories.COVER)]
         [Category(TestCategories.NORMAL)]
         public async Task NoRandomNumberGenerator01()
         {
-            var dist = new FastRng.Double.Distributions.StudentT();
+            var dist = new FastRng.Double.Distributions.StudentTNu1();
             Assert.DoesNotThrowAsync(async () => await dist.GetDistributedValue());
             Assert.That(await dist.GetDistributedValue(), Is.NaN);
         }