Interface Class for One Dimensional fitting in a weighted fashion. More...

#include <yat/regression/OneDimensionalWeighted.h>

Inheritance diagram for theplu::yat::regression::OneDimensionalWeighted:

Public Member Functions
	OneDimensionalWeighted (void)

virtual	~OneDimensionalWeighted (void)

virtual void	fit (const utility::VectorBase &x, const utility::VectorBase &y, const utility::VectorBase &w)=0

virtual double	predict (const double x) const =0

double	prediction_error2 (const double x, const double w=1.0) const

double	r2 (void) const

virtual double	s2 (double w=1) const =0

virtual double	standard_error2 (const double x) const =0

Protected Attributes
statistics::AveragerPairWeighted	ap_

double	chisq_
	Chi-squared. More...

Detailed Description

Interface Class for One Dimensional fitting in a weighted fashion.

Constructor & Destructor Documentation

theplu::yat::regression::OneDimensionalWeighted::OneDimensionalWeighted ( void )

Default Constructor.

virtual theplu::yat::regression::OneDimensionalWeighted::~OneDimensionalWeighted ( void )

virtual

Destructor

Member Function Documentation

virtual void theplu::yat::regression::OneDimensionalWeighted::fit	(	const utility::VectorBase &	x,
		const utility::VectorBase &	y,
		const utility::VectorBase &	w
	)

pure virtual

This function computes the best-fit given a model (see specific class for details) by minimizing $\sum{w_i(\hat{y_i}-y_i)^2}$ , where $\hat{y}$ is the fitted value. The weight $ w_i $ should be proportional to the inverse of the variance for $ y_i $

Implemented in theplu::yat::regression::LinearWeighted, theplu::yat::regression::NaiveWeighted, and theplu::yat::regression::PolynomialWeighted.

virtual double theplu::yat::regression::OneDimensionalWeighted::predict ( const double x ) const

pure virtual

Returns: expected value in x according to the fitted model

Implemented in theplu::yat::regression::LinearWeighted, theplu::yat::regression::PolynomialWeighted, and theplu::yat::regression::NaiveWeighted.

double theplu::yat::regression::OneDimensionalWeighted::prediction_error2	(	const double	x,
		const double	w = `1.0`
	)		const

The prediction error is defined as expected squared deviation a new data point (with weight w) will be from the model value $E((Y|x - \hat{y}(x))^2|w)$ and is typically divided into the conditional variance ( see s2() ) given $ x $ and the squared standard error ( see standard_error2() ) of the model estimation in $ x $ .