bdm::merger_base Class Reference

Base class for general combination of pdfs on discrete support. More...

Detailed Description

Base class for general combination of pdfs on discrete support.

Mixtures of Gaussian densities are used internally. Switching to other densities should be trivial.

The merged pdfs are expected to be of the form:

$f(x_i|y_i), i=1..n$

where the resulting merger is a density on $\cup [x_i,y_i]$ . Note that all variables will be joined.

As a result of this feature, each source must be extended to common support

$f(z_i|y_i,x_i) f(x_i|y_i) f(y_i) i=1..n$

where $ z_i $ accumulate variables that were not in the original source. These extensions are calculated on-the-fly.

However, these operations can not be performed in general. Hence, this class merges only sources on common support, $y_i={}, z_i={}, \forall i$ . For merging of more general cases, use offsprings merger_mix and merger_grid.

#include <merger.h>

List of all members.

Public Member Functions
virtual string	to_string ()
	This method returns a basic info about the current instance.
virtual void	to_setting (Setting &set) const
	This method save all the instance properties into the Setting structure.
Constructors

	merger_base ()
	Default constructor.
	merger_base (const Array< shared_ptr< mpdf > > &S)
	Constructor from sources.
void	set_sources (const Array< shared_ptr< mpdf > > &Sources)
	Function setting the main internal structures.
void	set_support (rectangular_support &Sup)
	Set support points from rectangular grid.
void	set_support (discrete_support &Sup)
	Set support points from dicrete grid.
void	set_debug_file (const string fname)
	set debug file
void	set_method (MERGER_METHOD MTH=DFLT_METHOD, double beta0=DFLT_beta)
	Set internal parameters used in approximation.
void	set_support (const epdf &overall, int N)
	Set support points from a pdf by drawing N samples.
virtual	~merger_base ()
	Destructor.
Mathematical operations

virtual void	merge ()
	Merge given sources in given points.
vec	merge_points (mat &lW)
	Merge log-likelihood values in points using method specified by parameter METHOD.
vec	mean () const
	weight w is a
mat	covariance () const
vec	variance () const
	return expected variance (not covariance!)
Access to attributes

eEmp &	_Smp ()
	Access function.
void	from_setting (const Setting &set)
	load from setting
void	validate ()
	This method TODO.
Constructors
Construction of each epdf should support two types of constructors: empty constructor, copy constructor, The following constructors should be supported for convenience: constructor followed by calling `set_parameters()` constructor accepting random variables calling `set_rv()` All internal data structures are constructed as empty. Their values (including sizes) will be set by method `set_parameters()`. This way references can be initialized in constructors.
void	set_parameters (int dim0)
Matematical Operations

virtual vec	sample () const
	Returns a sample, from density .
virtual mat	sample_m (int N) const
	Returns N samples, $[x_1 , x_2 , \ldots \$ from density .
virtual double	evallog (const vec &val) const
	Compute log-probability of argument `val` In case the argument is out of suport return -Infinity.
virtual vec	evallog_m (const mat &Val) const
	Compute log-probability of multiple values argument `val`.
virtual vec	evallog_m (const Array< vec > &Avec) const
	Compute log-probability of multiple values argument `val`.
virtual shared_ptr< mpdf >	condition (const RV &rv) const
	Return conditional density on the given RV, the remaining rvs will be in conditioning.
virtual shared_ptr< epdf >	marginal (const RV &rv) const
	Return marginal density on the given RV, the remainig rvs are intergrated out.
virtual void	qbounds (vec &lb, vec &ub, double percentage=0.95) const
	Lower and upper bounds of `percentage` % quantile, returns mean-2*sigma as default.
Connection to other classes
Description of the random quantity via attribute `rv` is optional. For operations such as sampling `rv` does not need to be set. However, for `marginalization` and `conditioning` `rv` has to be set. NB:
void	set_rv (const RV &rv0)
	Name its rv.
bool	isnamed () const
	True if rv is assigned.
const RV &	_rv () const
	Return name (fails when isnamed is false).
Access to attributes

int	dimension () const
	Size of the random variable.
Protected Attributes
Array< shared_ptr< mpdf > >	mpdfs
	Elements of composition.
Array< datalink_m2e * >	dls
	Data link for each mpdf in mpdfs.
Array< RV >	rvzs
	Array of rvs that are not modelled by mpdfs at all, .
Array< datalink_m2e * >	zdls
	Data Links for extension .
int	Npoints
	number of support points
int	Nsources
	number of sources
MERGER_METHOD	METHOD
	switch of the methoh used for merging
double	beta
	Prior on the log-normal merging model.
eEmp	eSmp
	Projection to empirical density (could also be piece-wise linear).
bool	DBG
	debug or not debug
it_file *	dbg_file
	debugging file
int	dim
	dimension of the random variable
RV	rv
	Description of the random variable.
Static Protected Attributes
static const MERGER_METHOD	DFLT_METHOD = LOGNORMAL
	Default for METHOD.
static const double	DFLT_beta = 1.2
	default for beta