"Random effect" with parameter expansion More...

#include <MuGen.h>

Inheritance diagram for MuGrpPEX:

Collaboration diagram for MuGrpPEX:

Public Member Functions
	MuGrpPEX ()
	Default constructor.

	MuGrpPEX (const Grp &dat, RanIndex &low, RanIndex &up, const double &Spr, const int &nThr)
	Full constructor. More...

	MuGrpPEX (const Grp &dat, RanIndex &low, RanIndex &up, const string outMuFlNam, const double &Spr, const int &nThr)
	Full constructor with location parameter file name. More...

	~MuGrpPEX ()
	Destructor.

	MuGrpPEX (const MuGrpPEX &mGp)
	Copy constructor. More...

MuGrpPEX &	operator= (const MuGrpPEX &mGp)
	Assignment operator. More...

const gsl_matrix *	fMat () const
	Access the adjusted value matrix. More...

void	save ()
	Save the adjusted values. More...

void	save (const string &outFlNam)
	Save adjusted values to named file. More...

void	save (const SigmaI &SigI)
	Save with the covariance matrix. More...

Apex &	getA ()
	Access the redundant parameter. More...

void	setApr (const double &pr)
	Set the inverse-prior for the redundant parameter matrix. More...

MuGrp	mean (RanIndex &grp)
	Group mean. More...

const MuGrp	mean (RanIndex &grp) const
	Group mean. More...

MuGrp	mean (RanIndex &grp, const Qgrp &q)
	Group weighted mean. More...

const MuGrp	mean (RanIndex &grp, const Qgrp &q) const
	Group weighted mean. More...

void	update (const Grp &dat, const SigmaI &SigIm, const SigmaI &SigIp)
	Gaussian likelihood, 0-mean Gaussian prior. More...

void	update (const Grp &dat, const Qgrp &q, const SigmaI &SigIm, const SigmaI &SigIp)
	Student- \(t\) likelihood, 0-mean Gaussian prior. More...

void	update (const Grp &dat, const SigmaI &SigIm, const Qgrp &qPr, const SigmaI &SigIp)
	Gaussian likelihood, 0-mean Student- \(t\) prior. More...

void	update (const Grp &dat, const Qgrp &q, const SigmaI &SigIm, const Qgrp &qPr, const SigmaI &SigIp)
	Student- \(t\) likelihood, 0-mean Student- \(t\) prior. More...

void	update (const Grp &dat, const SigmaI &SigIm, const Grp &muPr, const SigmaI &SigIp)
	Gaussian likelihood, non-zero mean Gaussian prior. More...

void	update (const Grp &dat, const Qgrp &q, const SigmaI &SigIm, const Grp &muPr, const SigmaI &SigIp)
	Student- \(t\) likelihood, non-zero mean Gaussian prior. More...

void	update (const Grp &dat, const SigmaI &SigIm, const Grp &muPr, const Qgrp &qPr, const SigmaI &SigIp)
	Gaussian likelihood, non-zero mean Student- \(t\) prior. More...

void	update (const Grp &dat, const Qgrp &q, const SigmaI &SigIm, const Grp &muPr, const Qgrp &qPr, const SigmaI &SigIp)
	Student- \(t\) likelihood, non-zero mean Student- \(t\) prior. More...

Public Member Functions inherited from MuGrp
	MuGrp ()
	Default constructor.

	MuGrp (RanIndex &low, const size_t &d)
	Deterministic zero-value constructor. More...

	MuGrp (const string &datFlNam, RanIndex &low, RanIndex &up, const size_t &d)
	Constructor with data from file. More...

	MuGrp (const string &datFlNam, RanIndex &up, const size_t &d)
	Constructor with data from file and no lower level. More...

	MuGrp (const vector< MVnorm * > &dat, RanIndex &low, RanIndex &up)
	Constructor with a vector of MVnorm pointers. More...

	MuGrp (const Grp &dat, RanIndex &low, RanIndex &up)
	Constructor with a Grp object. More...

	MuGrp (const vector< MVnorm * > &dat, RanIndex &low, RanIndex &up, const string &outFlNam)
	Constructor with a vector of MVnorm pointers and output file name. More...

	MuGrp (const Grp &dat, RanIndex &low, RanIndex &up, const string &outFlNam)
	Constructor with a Grp object and output file name. More...

	MuGrp (const Grp &dat, RanIndex &low)
	Deterministic mean constructor. More...

	MuGrp (const Grp &dat, const Qgrp &q, RanIndex &low)
	Deterministic weighted mean constructor. More...

	MuGrp (const gsl_matrix *dat)
	Deterministic constructor with a GSL matrix. More...

	MuGrp (const gsl_matrix *dat, RanIndex &low)
	Deterministic GSL matrix mean constructor. More...

	MuGrp (const gsl_matrix *dat, const Qgrp &q, RanIndex &low)
	Deterministic GSL matrix weighted mean constructor. More...

virtual	~MuGrp ()
	Destructor.

	MuGrp (const MuGrp &mG)
	Copy constructor. More...

	MuGrp (const Grp &g)
	Copy constructor. More...

MuGrp &	operator= (const MuGrp &mG)
	Assignemnt operator. More...

virtual void	update (const Grp &dat, const SigmaI &SigIm)
	Gaussian likelihood, improper prior. More...

virtual void	update (const Grp &dat, const Qgrp &q, const SigmaI &SigIm)
	Student- \(t\) likelihood, improper prior. More...

Public Member Functions inherited from Grp
virtual	~Grp ()
	Destructor.

virtual void	save (const string &outMuFlNam, const string &outSigFlNam, const SigmaI &SigI)
	Joint save. More...

virtual void	save (const Grp &y, const SigmaI &SigI)
	Save with data and inverse-covariance. More...

void	mhlSave (const string &outFlNam, const SigmaI SigI)
	Save Mahalanobis distance. More...

virtual void	dump ()
	Dump to a file. More...

const vector< MVnorm * > &	dataVec () const
	Get vector of row pointers. More...

virtual const gsl_matrix *	dMat () const
	Access the value matrix. More...

const size_t	Ndata () const
	Get number of rows. More...

const size_t	phenD () const
	Get number of traits. More...

virtual double	lnOddsRat (const Grp &y, const SigmaI &SigI, const size_t i) const
	Log-odds ratio. More...

const MVnorm *	operator[] (const size_t i) const
	Subscript operator. More...

MVnorm *	operator[] (const size_t i)
	Subscript operator. More...

void	center ()
	Center the value matrix. More...

Protected Member Functions
void	_updateFitted ()
	Update adjusted values. More...

Protected Member Functions inherited from Grp
	Grp ()

Protected Attributes
gsl_matrix *	_adjValMat
	Adjusted value matrix. More...

gsl_matrix *	_tSigIAt
	Scaled inverse-covariance. More...

Apex	_A
	Redundant parameter matrix.

vector< vector< double > >	_ftA
	Fitted values. More...

int	_nThr
	number of threads

string	_outSigFlNam
	Covariance output file name. More...

Protected Attributes inherited from Grp
vector< MVnorm * >	_theta
	Vector of pointers to value rows. More...

gsl_matrix *	_valueMat
	Value matrix. More...

RanIndex *	_lowLevel
	Lower level index. More...

RanIndex *	_upLevel
	Upper level index. More...

vector< gsl_rng * >	_rV
	Vector of PNG pointers. More...

string	_outFlNam
	Name of the output file.

Detailed Description

"Random effect" with parameter expansion

An implementation of a multivariate extension (Greenberg, unpublished) of the multiplicative PEX scheme for "random effects" model [gelman07] [gelman08] . This object has to be outside the model hierarchy, i.e. have the same prior for all rows. Variables are divided into "raw" and "adjusted" as in [gelman07] . "Adjusted" variables are on the correct scale for interpretation. The "raw" value matrix is denoted \( \boldsymbol{\Xi} \) and the redundant parameter matrix (implemented in the Apex class) is \( \boldsymbol{A} \).

Constructor & Destructor Documentation

◆ MuGrpPEX() [1/3]

MuGrpPEX::MuGrpPEX	(	const Grp &	dat,
		RanIndex &	low,
		RanIndex &	up,
		const double &	Spr,
		const int &	nThr
	)

Full constructor.

Initializing constructor that provides starting values for the location parameter and redundant parameter matrices.

Parameters

[in]	Grp&	data
[in]	RanIndex&	index to the lower (data) level
[in]	RanIndex&	index to the upper (prior) level
[in]	double&	inverse-prior for the redundant parameter matrix
[in]	int&	number of threads

◆ MuGrpPEX() [2/3]

MuGrpPEX::MuGrpPEX	(	const Grp &	dat,
		RanIndex &	low,
		RanIndex &	up,
		const string	outMuFlNam,
		const double &	Spr,
		const int &	nThr
	)

Full constructor with location parameter file name.

Initializing constructor that provides starting values for the location parameter and redundant parameter matrices. Sets the name of the file where the adjusted location parameter chains will be saved.

Parameters

[in]	Grp&	data
[in]	RanIndex&	index to the lower (data) level
[in]	RanIndex&	index to the upper (prior) level
[in]	string&	output file name
[in]	double&	inverse-prior for the redundant parameter matrix
[in]	int&	number of threads

◆ MuGrpPEX() [3/3]

MuGrpPEX::MuGrpPEX ( const MuGrpPEX & mGp )

Copy constructor.

Parameters

[in] MuGrpPEX& object to be copied

Returns: MuGrpPEX object

Member Function Documentation

◆ fMat()

const gsl_matrix* MuGrpPEX::fMat ( ) const

inlinevirtual

Access the adjusted value matrix.

Returns: gsl_matrix* pointer to the adjusted value matrix

Reimplemented from Grp.

◆ getA()

Apex& MuGrpPEX::getA ( )

inline

Access the redundant parameter.

Returns: Apex& the redundant parameter matrix

◆ mean() [1/4]

MuGrp MuGrpPEX::mean ( RanIndex & grp )

virtual

Group mean.

Calculates within-group mean values of the value matrix (by row). The groups are defined by the upper-level of the given RanIndex variable. The lower-level number of elements of the index has to be the same as the number of rows in the current object's value matrix.

Parameters

[in] RanIndex& grouping index

Returns: MuGrp object with the matrix of means

Reimplemented from Grp.

◆ mean() [2/4]

const MuGrp MuGrpPEX::mean ( RanIndex & grp ) const

virtual

Group mean.

Calculates within-group mean values of the value matrix (by row). The groups are defined by the upper-level of the given RanIndex variable. The lower-level number of elements of the index has to be the same as the number of rows in the current object's value matrix.

Parameters

[in] RanIndex& grouping index

Returns: MuGrp object with the matrix of means

Reimplemented from Grp.

◆ mean() [3/4]

MuGrp MuGrpPEX::mean	(	RanIndex &	grp,
		const Qgrp &	q
	)

virtual

Group weighted mean.

Calculates within-group weighted mean values of the value matrix (by row). The groups are defined by the upper-level of the given RanIndex variable. The lower-level number of elements of the index has to be the same as the number of rows in the current object's value matrix. The weights are in the provided Qgrp object. Typically, they are from a Student- \(t\) model.

Parameters

[in]	RanIndex&	grouping index
[in]	Qgrp&	weights

Returns: MuGrp object with the matrix of means

Reimplemented from Grp.

◆ mean() [4/4]

const MuGrp MuGrpPEX::mean	(	RanIndex &	grp,
		const Qgrp &	q
	)		const

virtual

Group weighted mean.

Calculates within-group weighted mean values of the value matrix (by row). The groups are defined by the upper-level of the given RanIndex variable. The lower-level number of elements of the index has to be the same as the number of rows in the current object's value matrix. The weights are in the provided Qgrp object. Typically, they are from a Student- \(t\) model.

Parameters

[in]	RanIndex&	grouping index
[in]	Qgrp&	weights

Returns: MuGrp object with the matrix of means

Reimplemented from Grp.

◆ operator=()

MuGrpPEX & MuGrpPEX::operator= ( const MuGrpPEX & mGp )

Assignment operator.

Parameters

[in] MuGrpPEX& object to be copied

Returns: MuGrpPEX& target object

◆ save() [1/3]

void MuGrpPEX::save ( )

virtual

Save the adjusted values.

Appends the current adjusted mean values to the file whose name was set during construction.

Reimplemented from Grp.

◆ save() [2/3]

void MuGrpPEX::save ( const SigmaI & SigI )

virtual

Save with the covariance matrix.

Appends the current adjusted mean values and prior covariance matrix to the files whose names were set during construction.

Reimplemented from Grp.

◆ save() [3/3]

void MuGrpPEX::save ( const string & outFlNam )

virtual

Save adjusted values to named file.

Appends the current adjusted mean values to the named file.

Parameters

[in] string& file name

Reimplemented from Grp.

◆ setApr()

void MuGrpPEX::setApr ( const double & pr )

inline

Set the inverse-prior for the redundant parameter matrix.

Parameters

[in] double& inverse-prior value

◆ update() [1/8]

void MuGrpPEX::update	(	const Grp &	,
		const Qgrp &	,
		const SigmaI &	,
		const Grp &	,
		const Qgrp &	,
		const SigmaI &
	)

virtual

Student- \(t\) likelihood, non-zero mean Student- \(t\) prior.

For the relationship to work properly, the _upLevel index of the focal object has to point to the rows of the prior mean matrix. This is the matrix addressed by dMat(). The relationship to the data is dependent on the derived class.

Parameters

[in]	Grp&	data
[in]	Qgrp&	Student- \(t\) weight parameter for the data
[in]	SigmaI&	data inverse-covariance
[in]	Grp&	prior mean
[in]	Qgrp&	Student- \(t\) weight parameter for the prior
[in]	SigmaI&	prior inverse-covariance

Reimplemented from MuGrp.

◆ update() [2/8]

void MuGrpPEX::update	(	const Grp &	,
		const Qgrp &	,
		const SigmaI &	,
		const Grp &	,
		const SigmaI &
	)

virtual

Student- \(t\) likelihood, non-zero mean Gaussian prior.

For the relationship to work properly, the _upLevel index of the focal object has to point to the rows of the prior mean matrix. This is the matrix addressed by dMat(). The relationship to the data is dependent on the derived class.

Parameters

[in]	Grp&	data
[in]	Qgrp&	Student- \(t\) weight parameter for the data
[in]	SigmaI&	data inverse-covariance
[in]	Grp&	prior mean
[in]	SigmaI&	prior inverse-covariance

Reimplemented from MuGrp.

◆ update() [3/8]

void MuGrpPEX::update	(	const Grp &	,
		const Qgrp &	,
		const SigmaI &	,
		const Qgrp &	,
		const SigmaI &
	)

virtual

Student- \(t\) likelihood, 0-mean Student- \(t\) prior.

Parameters

[in]	Grp&	data
[in]	Qgrp&	Student- \(t\) weight parameter for the data
[in]	SigmaI&	data inverse-covariance
[in]	Qgrp&	Student- \(t\) weight parameter for the prior
[in]	SigmaI&	prior inverse-covariance

Reimplemented from MuGrp.

◆ update() [4/8]

void MuGrpPEX::update	(	const Grp &	,
		const Qgrp &	,
		const SigmaI &	,
		const SigmaI &
	)

virtual

Student- \(t\) likelihood, 0-mean Gaussian prior.

Parameters

[in]	Grp&	data
[in]	Qgrp&	Student- \(t\) weight parameter for data
[in]	SigmaI&	data inverse-covariance
[in]	SigmaI&	prior inverse-covariance

Reimplemented from MuGrp.

◆ update() [5/8]

void MuGrpPEX::update	(	const Grp &	,
		const SigmaI &	,
		const Grp &	,
		const Qgrp &	,
		const SigmaI &
	)

virtual

Gaussian likelihood, non-zero mean Student- \(t\) prior.

For the relationship to work properly, the _upLevel index of the focal object has to point to the rows of the prior mean matrix. This is the matrix addressed by dMat(). The relationship to the data is dependent on the derived class.

Parameters

[in]	Grp&	data
[in]	SigmaI&	data inverse-covariance
[in]	Grp&	prior mean
[in]	Qgrp&	Student- \(t\) weight parameter for the prior
[in]	SigmaI&	prior inverse-covariance

Reimplemented from MuGrp.

◆ update() [6/8]

void MuGrpPEX::update	(	const Grp &	,
		const SigmaI &	,
		const Grp &	,
		const SigmaI &
	)

virtual

Gaussian likelihood, non-zero mean Gaussian prior.

For the relationship to work properly, the _upLevel index of the focal object has to point to the rows of the prior mean matrix. This is the matrix addressed by dMat(). The relationship to the data is dependent on the derived class.

Parameters

[in]	Grp&	data
[in]	SigmaI&	data inverse-covariance
[in]	Grp&	prior mean
[in]	SigmaI&	prior inverse-covariance

Reimplemented from MuGrp.

◆ update() [7/8]

void MuGrpPEX::update	(	const Grp &	,
		const SigmaI &	,
		const Qgrp &	,
		const SigmaI &
	)

virtual

Gaussian likelihood, 0-mean Student- \(t\) prior.

Parameters

[in]	Grp&	data
[in]	SigmaI&	data inverse-covariance
[in]	Qgrp&	Student- \(t\) weight parameter for the prior
[in]	SigmaI&	prior inverse-covariance

Reimplemented from MuGrp.

◆ update() [8/8]

void MuGrpPEX::update	(	const Grp &	,
		const SigmaI &	,
		const SigmaI &
	)

virtual

Gaussian likelihood, 0-mean Gaussian prior.

Parameters

[in]	Grp&	data
[in]	SigmaI&	data inverse-covariance
[in]	SigmaI&	prior inverse-covariance

Reimplemented from MuGrp.

Member Data Documentation

◆ _adjValMat

gsl_matrix* MuGrpPEX::_adjValMat

protected

Adjusted value matrix.

The location paramter of interest, i.e. \( \boldsymbol{\Xi A} \).

◆ _ftA

vector<vector<double> > MuGrpPEX::_ftA

protected

Fitted values.

Vector of vectorized individual fitted matrices \( \boldsymbol{\Xi}_{\cdot -m}\boldsymbol{A}_{-m\cdot} \).

◆ _outSigFlNam

string MuGrpPEX::_outSigFlNam

protected

Covariance output file name.

This is to save the adjusted prior covariance matrix associated with this object. Saving from the SigmaI object gives the raw matrix.

◆ _tSigIAt

gsl_matrix* MuGrpPEX::_tSigIAt

protected

Scaled inverse-covariance.

The matrix \( \left( \boldsymbol{\Sigma}^{-1}\boldsymbol{A}^T \right)^T \) that is common for sampling each row of the value matrix.

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

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ MuGrpPEX() [1/3]

◆ MuGrpPEX() [2/3]

◆ MuGrpPEX() [3/3]

Member Function Documentation

◆ fMat()

◆ getA()

◆ mean() [1/4]

◆ mean() [2/4]

◆ mean() [3/4]

◆ mean() [4/4]

◆ operator=()

◆ save() [1/3]

◆ save() [2/3]

◆ save() [3/3]

◆ setApr()

◆ update() [1/8]

◆ update() [2/8]

◆ update() [3/8]

◆ update() [4/8]

◆ update() [5/8]

◆ update() [6/8]

◆ update() [7/8]

◆ update() [8/8]

Member Data Documentation

◆ _adjValMat

◆ _ftA

◆ _outSigFlNam

◆ _tSigIAt