|
|
|
@ -57,25 +57,11 @@ namespace MathNet.Numerics.Distributions |
|
|
|
/// to <c>false</c>, all parameter checks can be turned off.</para></remarks>
|
|
|
|
public class StudentT : IContinuousDistribution |
|
|
|
{ |
|
|
|
/// <summary>
|
|
|
|
/// Keeps track of the location of the Student t-distribution.
|
|
|
|
/// </summary>
|
|
|
|
double _location; |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
/// Keeps track of the degrees of freedom for the Student t-distribution.
|
|
|
|
/// </summary>
|
|
|
|
double _dof; |
|
|
|
System.Random _random; |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
/// Keeps track of the scale for the Student t-distribution.
|
|
|
|
/// </summary>
|
|
|
|
double _location; |
|
|
|
double _scale; |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
/// The distribution's random number generator.
|
|
|
|
/// </summary>
|
|
|
|
System.Random _random; |
|
|
|
double _freedom; |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
/// Initializes a new instance of the StudentT class. This is a Student t-distribution with location 0.0
|
|
|
|
@ -122,7 +108,7 @@ namespace MathNet.Numerics.Distributions |
|
|
|
/// <returns>a string representation of the distribution.</returns>
|
|
|
|
public override string ToString() |
|
|
|
{ |
|
|
|
return "StudentT(Location = " + _location + ", Scale = " + _scale + ", DoF = " + _dof + ")"; |
|
|
|
return "StudentT(Location = " + _location + ", Scale = " + _scale + ", DoF = " + _freedom + ")"; |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
@ -153,7 +139,16 @@ namespace MathNet.Numerics.Distributions |
|
|
|
|
|
|
|
_location = location; |
|
|
|
_scale = scale; |
|
|
|
_dof = dof; |
|
|
|
_freedom = dof; |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
/// Gets or sets the random number generator which is used to draw random samples.
|
|
|
|
/// </summary>
|
|
|
|
public System.Random RandomSource |
|
|
|
{ |
|
|
|
get { return _random; } |
|
|
|
set { _random = value ?? new System.Random(); } |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
@ -162,7 +157,7 @@ namespace MathNet.Numerics.Distributions |
|
|
|
public double Location |
|
|
|
{ |
|
|
|
get { return _location; } |
|
|
|
set { SetParameters(value, _scale, _dof); } |
|
|
|
set { SetParameters(value, _scale, _freedom); } |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
@ -171,7 +166,7 @@ namespace MathNet.Numerics.Distributions |
|
|
|
public double Scale |
|
|
|
{ |
|
|
|
get { return _scale; } |
|
|
|
set { SetParameters(_location, value, _dof); } |
|
|
|
set { SetParameters(_location, value, _freedom); } |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
@ -179,33 +174,16 @@ namespace MathNet.Numerics.Distributions |
|
|
|
/// </summary>
|
|
|
|
public double DegreesOfFreedom |
|
|
|
{ |
|
|
|
get { return _dof; } |
|
|
|
get { return _freedom; } |
|
|
|
set { SetParameters(_location, _scale, value); } |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
/// Gets or sets the random number generator which is used to draw random samples.
|
|
|
|
/// </summary>
|
|
|
|
public System.Random RandomSource |
|
|
|
{ |
|
|
|
get { return _random; } |
|
|
|
set |
|
|
|
{ |
|
|
|
if (value == null) |
|
|
|
{ |
|
|
|
throw new ArgumentNullException(); |
|
|
|
} |
|
|
|
|
|
|
|
_random = value; |
|
|
|
} |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
/// Gets the mean of the Student t-distribution.
|
|
|
|
/// </summary>
|
|
|
|
public double Mean |
|
|
|
{ |
|
|
|
get { return _dof > 1.0 ? _location : Double.NaN; } |
|
|
|
get { return _freedom > 1.0 ? _location : Double.NaN; } |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
@ -215,17 +193,17 @@ namespace MathNet.Numerics.Distributions |
|
|
|
{ |
|
|
|
get |
|
|
|
{ |
|
|
|
if (Double.IsPositiveInfinity(_dof)) |
|
|
|
if (Double.IsPositiveInfinity(_freedom)) |
|
|
|
{ |
|
|
|
return _scale*_scale; |
|
|
|
} |
|
|
|
|
|
|
|
if (_dof > 2.0) |
|
|
|
if (_freedom > 2.0) |
|
|
|
{ |
|
|
|
return _dof*_scale*_scale/(_dof - 2.0); |
|
|
|
return _freedom*_scale*_scale/(_freedom - 2.0); |
|
|
|
} |
|
|
|
|
|
|
|
return _dof > 1.0 ? Double.PositiveInfinity : Double.NaN; |
|
|
|
return _freedom > 1.0 ? Double.PositiveInfinity : Double.NaN; |
|
|
|
} |
|
|
|
} |
|
|
|
|
|
|
|
@ -236,17 +214,17 @@ namespace MathNet.Numerics.Distributions |
|
|
|
{ |
|
|
|
get |
|
|
|
{ |
|
|
|
if (Double.IsPositiveInfinity(_dof)) |
|
|
|
if (Double.IsPositiveInfinity(_freedom)) |
|
|
|
{ |
|
|
|
return Math.Sqrt(_scale*_scale); |
|
|
|
} |
|
|
|
|
|
|
|
if (_dof > 2.0) |
|
|
|
if (_freedom > 2.0) |
|
|
|
{ |
|
|
|
return Math.Sqrt(_dof*_scale*_scale/(_dof - 2.0)); |
|
|
|
return Math.Sqrt(_freedom*_scale*_scale/(_freedom - 2.0)); |
|
|
|
} |
|
|
|
|
|
|
|
return _dof > 1.0 ? Double.PositiveInfinity : Double.NaN; |
|
|
|
return _freedom > 1.0 ? Double.PositiveInfinity : Double.NaN; |
|
|
|
} |
|
|
|
} |
|
|
|
|
|
|
|
@ -262,7 +240,7 @@ namespace MathNet.Numerics.Distributions |
|
|
|
throw new NotSupportedException(); |
|
|
|
} |
|
|
|
|
|
|
|
return (((_dof + 1.0)/2.0)*(SpecialFunctions.DiGamma((1.0 + _dof)/2.0) - SpecialFunctions.DiGamma(_dof/2.0))) + Math.Log(Math.Sqrt(_dof)*SpecialFunctions.Beta(_dof/2.0, 1.0/2.0)); |
|
|
|
return (((_freedom + 1.0)/2.0)*(SpecialFunctions.DiGamma((1.0 + _freedom)/2.0) - SpecialFunctions.DiGamma(_freedom/2.0))) + Math.Log(Math.Sqrt(_freedom)*SpecialFunctions.Beta(_freedom/2.0, 1.0/2.0)); |
|
|
|
} |
|
|
|
} |
|
|
|
|
|
|
|
@ -273,7 +251,7 @@ namespace MathNet.Numerics.Distributions |
|
|
|
{ |
|
|
|
get |
|
|
|
{ |
|
|
|
if (_dof <= 3) |
|
|
|
if (_freedom <= 3) |
|
|
|
{ |
|
|
|
throw new NotSupportedException(); |
|
|
|
} |
|
|
|
@ -322,15 +300,15 @@ namespace MathNet.Numerics.Distributions |
|
|
|
public double Density(double x) |
|
|
|
{ |
|
|
|
// TODO JVG we can probably do a better job for Cauchy special case
|
|
|
|
if (_dof >= 1e+8d) |
|
|
|
if (_freedom >= 1e+8d) |
|
|
|
{ |
|
|
|
return Normal.Density(_location, _scale, x); |
|
|
|
} |
|
|
|
|
|
|
|
var d = (x - _location)/_scale; |
|
|
|
return Math.Exp(SpecialFunctions.GammaLn((_dof + 1.0)/2.0) - SpecialFunctions.GammaLn(_dof/2.0)) |
|
|
|
*Math.Pow(1.0 + (d*d/_dof), -0.5*(_dof + 1.0)) |
|
|
|
/Math.Sqrt(_dof*Math.PI) |
|
|
|
return Math.Exp(SpecialFunctions.GammaLn((_freedom + 1.0)/2.0) - SpecialFunctions.GammaLn(_freedom/2.0)) |
|
|
|
*Math.Pow(1.0 + (d*d/_freedom), -0.5*(_freedom + 1.0)) |
|
|
|
/Math.Sqrt(_freedom*Math.PI) |
|
|
|
/_scale; |
|
|
|
} |
|
|
|
|
|
|
|
@ -342,16 +320,16 @@ namespace MathNet.Numerics.Distributions |
|
|
|
public double DensityLn(double x) |
|
|
|
{ |
|
|
|
// TODO JVG we can probably do a better job for Cauchy special case
|
|
|
|
if (_dof >= 1e+8d) |
|
|
|
if (_freedom >= 1e+8d) |
|
|
|
{ |
|
|
|
return Normal.DensityLn(_location, _scale, x); |
|
|
|
} |
|
|
|
|
|
|
|
var d = (x - _location)/_scale; |
|
|
|
return SpecialFunctions.GammaLn((_dof + 1.0)/2.0) |
|
|
|
- (0.5*((_dof + 1.0)*Math.Log(1.0 + (d*d/_dof)))) |
|
|
|
- SpecialFunctions.GammaLn(_dof/2.0) |
|
|
|
- (0.5*Math.Log(_dof*Math.PI)) - Math.Log(_scale); |
|
|
|
return SpecialFunctions.GammaLn((_freedom + 1.0)/2.0) |
|
|
|
- (0.5*((_freedom + 1.0)*Math.Log(1.0 + (d*d/_freedom)))) |
|
|
|
- SpecialFunctions.GammaLn(_freedom/2.0) |
|
|
|
- (0.5*Math.Log(_freedom*Math.PI)) - Math.Log(_scale); |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
@ -362,14 +340,14 @@ namespace MathNet.Numerics.Distributions |
|
|
|
public double CumulativeDistribution(double x) |
|
|
|
{ |
|
|
|
// TODO JVG we can probably do a better job for Cauchy special case
|
|
|
|
if (Double.IsPositiveInfinity(_dof)) |
|
|
|
if (Double.IsPositiveInfinity(_freedom)) |
|
|
|
{ |
|
|
|
return Normal.CumulativeDistribution(_location, _scale, x); |
|
|
|
} |
|
|
|
|
|
|
|
var k = (x - _location)/_scale; |
|
|
|
var h = _dof/(_dof + (k*k)); |
|
|
|
var ib = 0.5*SpecialFunctions.BetaRegularized(_dof/2.0, 0.5, h); |
|
|
|
var h = _freedom/(_freedom + (k*k)); |
|
|
|
var ib = 0.5*SpecialFunctions.BetaRegularized(_freedom/2.0, 0.5, h); |
|
|
|
return x <= _location ? ib : 1.0 - ib; |
|
|
|
} |
|
|
|
|
|
|
|
@ -396,7 +374,7 @@ namespace MathNet.Numerics.Distributions |
|
|
|
/// <returns>a sample from the distribution.</returns>
|
|
|
|
public double Sample() |
|
|
|
{ |
|
|
|
return SampleUnchecked(RandomSource, _location, _scale, _dof); |
|
|
|
return SampleUnchecked(RandomSource, _location, _scale, _freedom); |
|
|
|
} |
|
|
|
|
|
|
|
/// <summary>
|
|
|
|
@ -407,7 +385,7 @@ namespace MathNet.Numerics.Distributions |
|
|
|
{ |
|
|
|
while (true) |
|
|
|
{ |
|
|
|
yield return SampleUnchecked(RandomSource, _location, _scale, _dof); |
|
|
|
yield return SampleUnchecked(RandomSource, _location, _scale, _freedom); |
|
|
|
} |
|
|
|
} |
|
|
|
|
|
|
|
|
Christoph Ruegg
13 years ago