#include "math/stats/glm.h"

Inheritance diagram for MR::Math::Stats::GLM::TestFixedHeteroscedastic:

Protected Attributes
const index_array_type &	VG

const size_t	num_vgs

vector< size_t >	inputs_per_vg

vector_type	Rnn_sums

vector_type	inv_Rnn_sums

vector_type	gamma_weights

Protected Attributes inherited from MR::Math::Stats::GLM::TestFixedHomoscedastic
vector< Hypothesis::Partition >	partitions

const matrix_type	pinvM

const matrix_type	Rm

vector< matrix_type >	XtX

vector< default_type >	one_over_dof

Protected Attributes inherited from MR::Math::Stats::GLM::TestBase
const matrix_type &	y

const matrix_type	M

const vector< Hypothesis > &	c

std::shared_ptr< Math::Zstatistic >	stat2z

Detailed Description

A class to compute statistics from heteroscedastic data using a fixed General Linear Model. This class produces a statistic per effect of interest: Aspin-Welsh v for contrast vectors (i.e. Rank(C) == 1), sqrt(Welsh's v^2) for Rank(C) > 1. It should be used in cases where:

The same design matrix is to be applied for all image elements being tested;
The input data are considered to be heteroscedastic; that is, the variance is not equivalent between all observations, but these can be placed into "variance groups", within which all observations can be considered to have the same variance.

Definition at line 315 of file glm.h.

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

core/math/stats/glm.h

Protected Attributes

Detailed Description