You are here: Start » Function Reference » Data Classification » Regression Analysis » LinearRegression_M

LinearRegression_M

Header:	AVL.h
Namespace:	avl
Module:	FoundationPro

Computes linear regression of given point set using selected M-estimator for outlier suppression.

Syntax

void avl::LinearRegression_M
(
	const atl::Array<float>& inYValues,
	atl::Optional<const atl::Array<float>&> inXValues,
	avl::MEstimator::Type inOutlierSuppression,
	float inClippingFactor,
	int inIterationCount,
	atl::Optional<const avl::LinearFunction&> inInitialLinearFunction,
	avl::LinearFunction& outLinearFunction,
	atl::Array<float>& outEstimatedValues,
	atl::Array<float>& outResiduals,
	atl::Optional<atl::Array<float>&> outYInliers = atl::NIL,
	atl::Optional<atl::Array<float>&> outXInliers = atl::NIL
)

Parameters

Name	Type	Range	Default	Description
inYValues	const Array<float>&			Sequence of ordinates
inXValues	Optional<const Array<float>&>		NIL	Sequence of abscissae, or {0, 1, 2, ...} by default
inOutlierSuppression	MEstimator::Type
inClippingFactor	float	0.675 - 6.0	2.5f	Multitude of standard deviation within which points are considered inliers
inIterationCount	int	0 -	5	Number of iterations of outlier suppressing algorithm
inInitialLinearFunction	Optional<const LinearFunction&>		NIL	Initial approximation of the output linear function (if available)
outLinearFunction	LinearFunction&			Linear function approximating the given point set
outEstimatedValues	Array<float>&			The result of application of the computed function to the X values
outResiduals	Array<float>&			Difference between an input Y value and the corresponding estimated value
outYInliers	Optional<Array<float>&>		NIL	Coordinate of the inlying points of the best line
outXInliers	Optional<Array<float>&>		NIL	Coordinate of the inlying points of the best line

Optional Outputs

The computation of following outputs can be switched off by passing value atl::NIL to these parameters: outYInliers, outXInliers.

Errors

List of possible exceptions:

Error type	Description
DomainError	Inconsistent size of arrays in LinearRegression_M.