bdm::mratio Class Reference

Class representing ratio of two densities which arise e.g. by applying the Bayes rule. It represents density in the form:

\[ f(rv|rvc) = \frac{f(rv,rvc)}{f(rvc)} \]

where $ f(rvc) = \int f(rv,rvc) d\ rv $. More...

#include <emix.h>

List of all members.

Public Member Functions

 mratio (const epdf *nom0, const RV &rv, bool copy=false)
double evallogcond (const vec &val, const vec &cond)
 Shortcut for conditioning and evaluation of the internal epdf. In some cases, this operation can be implemented efficiently.
void ownnom ()
 Object takes ownership of nom and will destroy it.
 ~mratio ()
 Default destructor.
void from_setting (const Setting &set)
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.
virtual void validate ()
 This method TODO.
Matematical operations
virtual vec samplecond (const vec &cond)
 Returns a sample from the density conditioned on cond, $x \sim epdf(rv|cond)$.
virtual mat samplecond_m (const vec &cond, int N)
 Returns.
virtual vec evallogcond_m (const mat &Dt, const vec &cond)
 Matrix version of evallogcond.
virtual vec evallogcond_m (const Array< vec > &Dt, const vec &cond)
 Array<vec> version of evallogcond.
Access to attributes
RV _rv () const
RV _rvc ()
int dimension () const
int dimensionc ()
Connection to other objects
void set_rvc (const RV &rvc0)
void set_rv (const RV &rv0)
bool isnamed ()

Protected Member Functions

void set_ep (epdf &iepdf)
 set internal pointer ep to point to given iepdf
void set_ep (epdf *iepdfp)
 set internal pointer ep to point to given iepdf

Protected Attributes

const epdfnom
 Nominator in the form of mpdf.
shared_ptr< epdfden
 Denominator in the form of epdf.
bool destroynom
 flag for destructor
datalink_m2e dl
 datalink between conditional and nom
epdf iepdf
 dummy epdf that stores only rv and dim
int dimc
 dimension of the condition
RV rvc
 random variable in condition


Detailed Description

Class representing ratio of two densities which arise e.g. by applying the Bayes rule. It represents density in the form:

\[ f(rv|rvc) = \frac{f(rv,rvc)}{f(rvc)} \]

where $ f(rvc) = \int f(rv,rvc) d\ rv $.

In particular this type of arise by conditioning of a mixture model.

At present the only supported operation is evallogcond().


Constructor & Destructor Documentation

bdm::mratio::mratio ( const epdf nom0,
const RV rv,
bool  copy = false 
) [inline]

Default constructor. By default, the given epdf is not copied! It is assumed that this function will be used only temporarily.

References bdm::RV::_dsize(), bdm::epdf::_rv(), den, destroynom, bdm::mpdf::dimc, dl, iepdf, bdm::RV::length(), bdm::epdf::marginal(), nom, bdm::mpdf::rvc, bdm::datalink_m2e::set_connection(), bdm::mpdf::set_ep(), bdm::epdf::set_parameters(), bdm::epdf::set_rv(), and bdm::RV::subt().


Member Function Documentation

void bdm::mpdf::from_setting ( const Setting &  set  )  [virtual, inherited]

Load from structure with elements:

 { class = "mpdf_offspring",
   rv = {class="RV", names=(...),}; // RV describing meaning of random variable
   rvc= {class="RV", names=(...),}; // RV describing meaning of random variable in condition
   // elements of offsprings
 }

Reimplemented from bdm::root.

Reimplemented in bdm::mepdf, bdm::mprod, bdm::mlnorm< sq_T, TEpdf >, bdm::mgnorm< sq_T >, bdm::mgamma, bdm::migamma_ref, bdm::mlognorm, bdm::mlnorm< chmat >, and bdm::mlnorm< ldmat, enorm >.

virtual vec bdm::mpdf::samplecond ( const vec &  cond  )  [inline, virtual, inherited]

mat bdm::mpdf::samplecond_m ( const vec &  cond,
int  N 
) [virtual, inherited]


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

Generated on Sun Aug 16 17:58:19 2009 for mixpp by  doxygen 1.5.8