Class SobolResult

Namespace

ChebyshevSharp

Assembly

ChebyshevSharp.dll

Result of SobolIndices() on Chebyshev interpolants: per-dimension Sobol sensitivity indices computed from spectral Chebyshev coefficients (no Monte Carlo).

public sealed record SobolResult : IEquatable<SobolResult>

Inheritance

object

SobolResult

Implements

IEquatable<SobolResult>

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.ReferenceEquals(object, object)

object.ToString()

Constructors

SobolResult(double[], double[], double)

Create a Sobol result from first-order and total-order index arrays.

public SobolResult(double[] FirstOrder, double[] TotalOrder, double Variance)

Parameters

FirstOrder double[]

First-order index per dimension. Sums to ≤ 1; sums to 1 for purely additive functions.

TotalOrder double[]

Total-order index per dimension. FirstOrder[d] ≤ TotalOrder[d] always.

Variance double

Total spectral variance.

Properties

FirstOrder

First-order index per dimension. Sums to ≤ 1; sums to 1 for purely additive functions.

public double[] FirstOrder { get; init; }

Property Value

double[]

TotalOrder

Total-order index per dimension. FirstOrder[d] ≤ TotalOrder[d] always.

public double[] TotalOrder { get; init; }

Property Value

double[]

Variance

Total spectral variance Σ_{α≠0} c_α² ‖T_α‖². When zero or at numerical noise level, the function is effectively constant and the indices are meaningless.

public double Variance { get; init; }

Property Value

double

Methods

Deconstruct(out double[], out double[], out double)

Deconstruct into first-order indices, total-order indices, and variance.

public void Deconstruct(out double[] FirstOrder, out double[] TotalOrder, out double Variance)

Parameters

FirstOrder double[]

TotalOrder double[]

Variance double