Class SamplerCauchy

Namespace

NanoQL.Experimental

Assembly

NanoQL.dll

A sampler from the Cauchy distribution, x + T·c: for lower dimensions it can be faster than the Gaussian sampler, especially with the Cauchy temperature: QuantLib's SamplerCauchy.

[Experimental("NQL0001")]
[WorkingType]
public sealed class SamplerCauchy : IAnnealingSampler<SamplerCauchy>

Inheritance

object

SamplerCauchy

Implements

IAnnealingSampler<SamplerCauchy>

Inherited Members

object.GetType()

object.ToString()

object.Equals(object)

object.Equals(object, object)

object.ReferenceEquals(object, object)

object.GetHashCode()

Remarks

Its draws are std::cauchy_distribution<Real>(0.0, 1.0) on a std::mt19937, as MSVC draws them, on the seed, which is required: QuantLib's default seeds from the clock. A working type.

Constructors

SamplerCauchy(long)

Initializes a new instance of the SamplerCauchy class.

public SamplerCauchy(long seed)

Parameters

seed long

The seed, in [1, 4294967295].

Exceptions

ArgumentOutOfRangeException

seed is outside [1, 4294967295].

Methods

Clone()

A copy, its state included: the copy's next draw equals this sampler's.

public SamplerCauchy Clone()

Returns

SamplerCauchy

The copy.

Sample(Vector, Vector, Vector)

Draws a new point (QuantLib's operator()).

public void Sample(Vector newPoint, Vector currentPoint, Vector temp)

Parameters

newPoint Vector

The new point, which it sets.

currentPoint Vector

The current point.

temp Vector

Each dimension's temperature.

Exceptions

ArgumentException

The vectors' sizes differ, with QuantLib's text.