bdm::eDirich Class Reference

Dirichlet posterior density. More...

#include <libEF.h>

Inheritance diagram for bdm::eDirich:

Inheritance graph
[legend]
Collaboration diagram for bdm::eDirich:

Collaboration graph
[legend]

List of all members.

Public Member Functions

 eDirich (const RV &rv, const vec &beta0)
 Default constructor.
 eDirich (const eDirich &D0)
 Copy constructor.
vec sample () const
 Returns a sample, $x$ from density $epdf(rv)$.
vec mean () const
 return expected value
vec variance () const
 return expected variance (not covariance!)
double evallog_nn (const vec &val) const
 In this instance, val is ...
double lognc () const
 logarithm of the normalizing constant, $\mathcal{I}$
vec & _beta ()
 access function
void set_parameters (const vec &beta0)
 Set internal parameters.
virtual void dupdate (mat &v)
 TODO decide if it is really needed.
virtual double evallog (const vec &val) const
 Evaluate normalized log-probability.
virtual vec evallog (const mat &Val) const
 Evaluate normalized log-probability for many samples.
virtual void pow (double p)
 Power of the density, used e.g. to flatten the density.
virtual mat sample_m (int N) const
 Returns N samples from density $epdf(rv)$.
virtual vec evallog_m (const mat &Val) const
 Compute log-probability of multiple values argument val.
virtual mpdfcondition (const RV &rv) const
 Return conditional density on the given RV, the remaining rvs will be in conditioning.
virtual epdfmarginal (const RV &rv) const
 Return marginal density on the given RV, the remainig rvs are intergrated out.
const RV_rv () const
 access function, possibly dangerous!
void _renewrv (const RV &in_rv)
 modifier function - useful when copying epdfs

Protected Attributes

vec beta
 sufficient statistics
double gamma
 speedup variable
RV rv
 Identified of the random variable.


Detailed Description

Dirichlet posterior density.

Continuous Dirichlet density of $n$-dimensional variable $x$

\[ f(x|\beta) = \frac{\Gamma[\gamma]}{\prod_{i=1}^{n}\Gamma(\beta_i)} \prod_{i=1}^{n}x_i^{\beta_i-1} \]

where $\gamma=\sum_i \beta_i$.


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

Generated on Fri Feb 6 15:29:54 2009 for mixpp by  doxygen 1.5.6