Multivariate.MultiLinearRegression Method (IReadOnlyList<Double>, IReadOnlyList<IReadOnlyList<Double>>)

Performs a multivariate linear regression on the listed columns.

Namespace: Meta.Numerics.Statistics
Assembly: Meta.Numerics (in Meta.Numerics.dll) Version: 4.1.4

Syntax

public static MultiLinearRegressionResult MultiLinearRegression(
	IReadOnlyList<double> yColumn,
	IReadOnlyList<IReadOnlyList<double>> xColumns
)

Public Shared Function MultiLinearRegression ( 
	yColumn As IReadOnlyList(Of Double),
	xColumns As IReadOnlyList(Of IReadOnlyList(Of Double))
) As MultiLinearRegressionResult

public:
static MultiLinearRegressionResult^ MultiLinearRegression(
	IReadOnlyList<double>^ yColumn, 
	IReadOnlyList<IReadOnlyList<double>^>^ xColumns
)

static member MultiLinearRegression : 
        yColumn : IReadOnlyList<float> * 
        xColumns : IReadOnlyList<IReadOnlyList<float>> -> MultiLinearRegressionResult

Parameters

yColumn: Type: System.Collections.Generic.IReadOnlyList<Double>
A column of dependent variable observations to be predicted.
xColumns: Type: System.Collections.Generic.IReadOnlyList<IReadOnlyList<Double>>
A list of columns of independent variable observations that serve as inputs to the regression function.

Return Value

Type: MultiLinearRegressionResult
A multi-linear fit.

Exceptions

Exception	Condition
NullReferenceException	yColumn is null, or xColumns is null, or one of the x-columns in the list is null.
DimensionMismatchException	Not all of the columns contain the same number of observations.
InsufficientDataException	There are too few observations to perform the fit.

Remarks

Note that the xColumns argument is column-oriented, i.e. it is a list of columns, not row-oriented, i.e. a list of rows. Either of these would match the method signature, but only column-oriented inputs will produce correct results.

Reference

Multivariate Class

MultiLinearRegression Overload

Meta.Numerics.Statistics Namespace