theplu::yat::regression::MultiDimensional Class Reference

MultiDimesional fitting. More...

#include <yat/regression/MultiDimensional.h>

List of all members.

Public Member Functions

 MultiDimensional (void)
 Default Constructor.
 ~MultiDimensional (void)
 Destructor.
const utility::Matrixcovariance (void) const
 covariance of parameters
void fit (const utility::Matrix &X, const utility::VectorBase &y)
 Function fitting parameters of the linear model by miminizing the quadratic deviation between model and data.
const utility::Vectorfit_parameters (void) const
double chisq (void) const
 Summed Squared Error.
double predict (const utility::VectorBase &x) const
double prediction_error2 (const utility::VectorBase &x) const
double standard_error2 (const utility::VectorBase &x) const


Detailed Description

MultiDimesional fitting.

Member Function Documentation

void theplu::yat::regression::MultiDimensional::fit ( const utility::Matrix X,
const utility::VectorBase y 
)

Function fitting parameters of the linear model by miminizing the quadratic deviation between model and data.

Number of rows in X must match size of y.

Exceptions:
A GSL_error exception is thrown if memory allocation fails or the underlying GSL calls fails (usually matrix dimension errors).

const utility::Vector& theplu::yat::regression::MultiDimensional::fit_parameters ( void   )  const

Returns:
parameters of the model

double theplu::yat::regression::MultiDimensional::predict ( const utility::VectorBase x  )  const

Returns:
value in x according to fitted model

double theplu::yat::regression::MultiDimensional::prediction_error2 ( const utility::VectorBase x  )  const

Returns:
expected squared prediction error for a new data point in x

double theplu::yat::regression::MultiDimensional::standard_error2 ( const utility::VectorBase x  )  const

Returns:
squared error of model value in x


The documentation for this class was generated from the following file:

Generated on Tue Jan 18 02:21:18 2011 for yat by  doxygen 1.5.5