eDirich Class Reference

Dirichlet posterior density. More...

#include <libEF.h>

Inheritance diagram for eDirich:

Collaboration diagram for eDirich:

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

double evalpdflog_nn (const vec &val) const

In this instance, val= [theta, r]. For multivariate instances, it is stored columnwise val = [theta_1 theta_2 ... r_1 r_2 ].

double lognc () const

logarithm of the normalizing constant, $\mathcal{I}$

vec & _beta ()

access function

virtual void dupdate (mat &v)

TODO decide if it is really needed.

virtual double evalpdflog (const vec &val) const

Evaluate normalized log-probability.

virtual vec evalpdflog (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 sampleN (int N) const

Returns N samples from density $epdf(rv)$ .

virtual double eval (const vec &val) const

Compute probability of argument val.

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

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:

work/git/mixpp/bdm/stat/libEF.h


Public Member Functions
	eDirich (const RV &rv, const vec &beta0)
	Default constructor.
	eDirich (const eDirich &D0)
	Copy constructor.
vec	sample () const
	Returns a sample, from density .
vec	mean () const
	return expected value
double	evalpdflog_nn (const vec &val) const
	In this instance, val= [theta, r]. For multivariate instances, it is stored columnwise val = [theta_1 theta_2 ... r_1 r_2 ].
double	lognc () const
	logarithm of the normalizing constant, $\mathcal{I}$
vec &	_beta ()
	access function
virtual void	dupdate (mat &v)
	TODO decide if it is really needed.
virtual double	evalpdflog (const vec &val) const
	Evaluate normalized log-probability.
virtual vec	evalpdflog (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	sampleN (int N) const
	Returns N samples from density .
virtual double	eval (const vec &val) const
	Compute probability of argument `val`.
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
RV	rv
	Identified of the random variable.