eprod Class Reference

Chain rule decomposition of epdf. More...

#include <emix.h>

Inheritance diagram for eprod:

Inheritance graph
[legend]
Collaboration diagram for eprod:

Collaboration graph
[legend]

List of all members.

Public Member Functions

 eprod (Array< epdf > Facs)
virtual vec sample () const =0
 Returns the required moment of the epdf.
virtual mat sampleN (int N) const
 Returns N samples from density $epdf(rv)$.
virtual double eval (const vec &val) const
 Compute probability of argument val.
virtual double evalpdflog (const vec &val) const =0
 Compute log-probability of argument val.
virtual vec mean () const =0
 return expected value
RV _rv () const
 access function

Protected Attributes

Array< epdf * > epdfs
Array< mpdf * > mpdfs
ivec sizes
RV rv
 Identified of the random variable.


Detailed Description

Chain rule decomposition of epdf.

Probability density in the form of Chain-rule decomposition: \[ f(x_1,x_2,x_3) = f(x_1|x_2,x_3)f(x_2,x_3)f(x_3) \] Note that


Member Function Documentation

virtual vec epdf::sample (  )  const [pure virtual, inherited]

Returns the required moment of the epdf.

Returns a sample, $x$ from density $epdf(rv)$

Implemented in emix, enorm< sq_T >, egiw, egamma, euni, eEmp, enorm< ldmat >, enorm< chmat >, and enorm< fsqmat >.

Referenced by mpdf::samplecond(), PF::set_est(), and eEmp::set_parameters().


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

Generated on Sat Aug 16 17:22:08 2008 for mixpp by  doxygen 1.5.6