Click or drag to resize
MultipleRegressionSvdT Method (T, T, Boolean)
Find the model parameters β such that their linear combination with all predictor-arrays in X become as close to their response in Y as possible, with least squares residuals. Uses a singular value decomposition and is therefore more numerically stable (especially if ill-conditioned) than the normal equations or QR but also slower.

Namespace: MathNet.Numerics.LinearRegression
Assembly: MathNet.Numerics (in MathNet.Numerics.dll) Version: 3.7
Syntax
C#
public static T[] Svd<T>(
	T[][] x,
	T[] y,
	bool intercept = false
)
where T : struct, new(), IEquatable<T>, IFormattable

Parameters

x
Type: T
List of predictor-arrays.
y
Type: T
List of responses
intercept (Optional)
Type: SystemBoolean
True if an intercept should be added as first artificial predictor value. Default = false.

Type Parameters

T

Return Value

Type: T
Best fitting list of model parameters β for each element in the predictor-arrays.
See Also