mixpp: bdm::NoiseParticleXY Class Reference

bdm::NoiseParticleXY Class Reference

#include <particles.h>

Inheritance diagram for bdm::NoiseParticleXY:

bdm::BM bdm::root List of all members.

Public Member Functions

 NoiseParticleXY (const NoiseParticleXY &m2)
const eprod_3posterior () const
void set_prior (const epdf *pdf0)
BM_copy () const
void bayes (const vec &dt, const vec &cond)
void from_setting (const Setting &set)
void to_setting (Setting &set) const
void validate ()

Protected Attributes

dirac est_emp
 discrte particle
shared_ptr< BMbmx
 internal Bayes Model
shared_ptr< BMbmy
bdm::NoiseParticleXY::eprod_3 est
 Internal class for custom posterior - product of empirical and exact part.
shared_ptr< fncg
 function transforming xt, ut -> x_t+1
shared_ptr< fnch
 function transforming xt,ut -> yt
RV rvx
RV rvxc
RV rvyc
datalink_part cond2g
 link from condition to f
datalink_part cond2h
 link from condition to h
datalink_part x2g
 link from xt to f
datalink_part x2h
 link from xt to h

Classes

class  eprod_3
 Internal class for custom posterior - product of empirical and exact part. More...

Detailed Description

Marginalized particle for state-space models with unknown parameters of distribuution of residues on $v_t$ and $ w_t $.

\begin{eqnarray*}{ x_t &=& g(x_{t-1}) + v_t,\\ y_t &= &h(x_t)+w_t, \end{eqnarray*}

This particle is a only a shell creating the residues calling internal estimator of their parameters. The internal estimator can be of any compatible type, e.g. ARX for Gaussian residues with unknown mean and variance.


The documentation for this class was generated from the following file:
Generated on 2 Dec 2013 for mixpp by  doxygen 1.4.7