Definitions for univariate probability mass functions. More...

#include <jwsc/config.h>
#include <stdlib.h>
#include <math.h>
#include <assert.h>
#include <inttypes.h>
#include "jwsc/base/limits.h"
#include "jwsc/prob/pdf.h"
#include "jwsc/prob/pmf.h"

Include dependency graph for pmf.c:

Go to the source code of this file.

Defines
#define	INV_E_SINGLE 0.3678794411714423f
	Single precision inverse of the number e.
#define	INV_E_DOUBLE 0.36787944117144232159552377016146
	Double precision inverse of the number e.
#define	ALMOST_ONE_SINGLE 0.9999999999999999
	Single precision value almost equal to one.
#define	ALMOST_ONE_DOUBLE 0.99999999999999999999999999999999
	Double precision value almost equal to one.
Functions
poission_pmf
Computes the Poission pmf at a point.
float	poisson_pmf_f (float lambda, uint32_t n)
	Computes the double precision Poisson pmf at a point.
double	poisson_pmf_d (double lambda, uint32_t n)
	Computes the double precision Poisson pmf at a point.
sample_poisson_pmf
Draws a random sample from the Poisson pmf.
uint32_t	sample_poisson_pmf_f (float lambda, uint32_t a, uint32_t b)
	Draws a single precision random sample from the Poisson pmf.
uint32_t	sample_poisson_pmf_d (double lambda, uint32_t a, uint32_t b)
	Draws a double precision random sample from the Poisson pmf.
bernoulli_pmf
Computes the Bernoulli pmf at a point.
float	bernoulli_pmf_f (float p, uint32_t n)
	Computes the single precision Bernoulli pmf at a point.
double	bernoulli_pmf_d (double p, uint32_t n)
	Computes the double precision Bernoulli pmf at a point.
sample_bernoulli_pmf
Draws a random sample from the Bernoulli pmf.
uint8_t	sample_bernoulli_pmf_f (float p)
	Draws a single precision random sample from the Bernoulli pmf.
uint8_t	sample_bernoulli_pmf_d (double p)
	Draws a double precision random sample from the Bernoulli pmf.
geometric_pmf
Computes the Geometric pmf at a point.
float	geometric_pmf_f (float p, uint32_t n)
	Computes the single precision Geometric pmf at a point.
double	geometric_pmf_d (double p, uint32_t n)
	Computes the double precision Geometric pmf at a point.
log_geometric_pmf
Computes the log of the Geometric pmf at a point.
float	log_geometric_pmf_f (float p, uint32_t n)
	Computes the single precision log Geometric pmf at a point.
double	log_geometric_pmf_d (double p, uint32_t n)
	Computes the double precision log Geometric pmf at a point.
sample_geometric_pmf
Draws a random sample from the Geometric pmf.
uint32_t	sample_geometric_pmf_f (float p)
	Draws a single precision random sample from the Geometric pmf.
uint32_t	sample_geometric_pmf_d (double p)
	Draws a double precision random sample from the Geometric pmf.

Detailed Description

Definitions for univariate probability mass functions.

Author:: Joseph Schlecht

License:: Creative Commons BY-NC-SA 3.0

Definition in file pmf.c.

Define Documentation

#define INV_E_SINGLE 0.3678794411714423f

Single precision inverse of the number e.

Definition at line 59 of file pmf.c.

#define INV_E_DOUBLE 0.36787944117144232159552377016146

Double precision inverse of the number e.

Definition at line 62 of file pmf.c.

#define ALMOST_ONE_SINGLE 0.9999999999999999

Single precision value almost equal to one.

Definition at line 65 of file pmf.c.

#define ALMOST_ONE_DOUBLE 0.99999999999999999999999999999999

Double precision value almost equal to one.

Definition at line 68 of file pmf.c.

Function Documentation

float poisson_pmf_f	(	float	lambda,
		uint32_t	n
	)

Computes the double precision Poisson pmf at a point.

Parameters:

lambda	Mean and variance of the Poisson.
n	Value to get the probability of.

Returns:: Probability of n according to the specified Poisson.

Warning:: The computation in this function takes O(n) multiplications so as to maintain numerical stability. If lambda <= 0 and n == 0, 1 is returned. Otherwise, if lambda <= 0 and n != 0, 0 is returned.

Definition at line 88 of file pmf.c.

double poisson_pmf_d	(	double	lambda,
		uint32_t	n
	)

Computes the double precision Poisson pmf at a point.

Parameters:

lambda	Mean and variance of the Poisson.
n	Value to get the probability of.

Returns:: Probability of n according to the specified Poisson.

Warning:: The computation in this function takes O(n) multiplications so as to maintain numerical stability. If lambda <= 0 and n == 0, 1 is returned. Otherwise, if lambda <= 0 and n != 0, 0 is returned.

Definition at line 146 of file pmf.c.

uint32_t sample_poisson_pmf_f	(	float	lambda,
		uint32_t	a,
		uint32_t	b
	)

Draws a single precision random sample from the Poisson pmf.

Parameters:

lambda	Mean and variance of the Poisson.
a	Lower bound of the sampling interval.
b	Upper bound of the sampling interval.

Returns:: A random sample from the Poisson distrition with parameter lambda over the integer interval [a,b].

Warning:: The computation in this function takes O(n) multiplications so as to maintain numerical stability. If lambda <= 0 and lambda is not in the range [a,b], a or b is returned, which ever is closest to lambda. Otherwise if lambda <= 0 and lambda is in the range [a,b], lambda is returned.

Note:: Does not seed the rand() function. If this is desired, srand() must be called prior to calling this function.

Definition at line 222 of file pmf.c.

uint32_t sample_poisson_pmf_d	(	double	lambda,
		uint32_t	a,
		uint32_t	b
	)

Draws a double precision random sample from the Poisson pmf.

Parameters:

lambda	Mean and variance of the Poisson.
a	Lower bound of the sampling interval.
b	Upper bound of the sampling interval.

Returns:: A random sample from the Poisson distrition with parameter lambda over the integer interval [a,b].

Warning:: The computation in this function takes O(n) multiplications so as to maintain numerical stability. If lambda <= 0 and lambda is not in the range [a,b], a or b is returned, which ever is closest to lambda. Otherwise if lambda <= 0 and lambda is in the range [a,b], lambda is returned.

Note:: Does not seed the rand() function. If this is desired, srand() must be called prior to calling this function.

Definition at line 281 of file pmf.c.

float bernoulli_pmf_f	(	float	p,
		uint32_t	n
	)

Computes the single precision Bernoulli pmf at a point.

Parameters:

p	Probability of success, i.e., n == 1.
n	Value to get the probability of. Must be either 0 or 1.

Returns:: Probability of n according to the specified Bernoulli.

Warning:: If n > 1, it is considered as 1.

Definition at line 340 of file pmf.c.

double bernoulli_pmf_d	(	double	p,
		uint32_t	n
	)

Computes the double precision Bernoulli pmf at a point.

Parameters:

p	Probability of success, i.e., n == 1.
n	Value to get the probability of. Must be either 0 or 1.

Returns:: Probability of n according to the specified Bernoulli.

Warning:: If n > 1, it is considered as 1.

Definition at line 353 of file pmf.c.

uint8_t sample_bernoulli_pmf_f ( float p )

Draws a single precision random sample from the Bernoulli pmf.

Parameters:

p	Probability of success.

Returns:: A random sample from the Bernoulli distrition with parameter p over the integer interval [0,1].

Note:: Does not seed the rand() function. If this is desired, srand() must be called prior to calling this function.

Definition at line 377 of file pmf.c.

uint8_t sample_bernoulli_pmf_d ( double p )

Draws a double precision random sample from the Bernoulli pmf.

Parameters:

p	Probability of success.

Returns:: A random sample from the Bernoulli distrition with parameter p over the integer interval [0,1].

Note:: Does not seed the rand() function. If this is desired, srand() must be called prior to calling this function.

Definition at line 402 of file pmf.c.

float geometric_pmf_f	(	float	p,
		uint32_t	n
	)

Computes the single precision Geometric pmf at a point.

Parameters:

p	Probability of success, must be in (0,1).
n	Number of failures before a success, must be >= 0.

Computes

 (1-p)^n * p

Note:: The value p must be in (0,1).

Returns:: Probability of n according to the specified Geometric.

Definition at line 440 of file pmf.c.

double geometric_pmf_d	(	double	p,
		uint32_t	n
	)

Computes the double precision Geometric pmf at a point.

Parameters:

p	Probability of success, must be in (0,1).
n	Number of failures before a success, must be >= 0.

Computes

 (1-p)^n * p

Note:: The value p must be in (0,1).

Returns:: Probability of n according to the specified Geometric.

Definition at line 469 of file pmf.c.

float log_geometric_pmf_f	(	float	p,
		uint32_t	n
	)

Computes the single precision log Geometric pmf at a point.

Parameters:

p	Probability of success, must be in (0,1).
n	Number of failures before a success, must be >= 0.

Computes

 log((1-p)^n * p)

Note:: The value p must be in (0,1).

Returns:: Probability of n according to the specified Geometric.

Definition at line 508 of file pmf.c.

double log_geometric_pmf_d	(	double	p,
		uint32_t	n
	)

Computes the double precision log Geometric pmf at a point.

Parameters:

p	Probability of success, must be in (0,1).
n	Number of failures before a success, must be >= 0.

Computes

 (1-p)^n * p

Note:: The value p must be in (0,1).

Returns:: Probability of n according to the specified Geometric.

Definition at line 537 of file pmf.c.

uint32_t sample_geometric_pmf_f ( float p )

Draws a single precision random sample from the Geometric pmf.

Parameters:

p	Probability of success, must be in (0,1).

Returns:: A random sample from the geometric distribution with parameter p.

Warning:: The parameter p must be >= 0.

Note:: Does not seed the rand() function. If this is desired, srand() must be called prior to calling this function.

Definition at line 574 of file pmf.c.

uint32_t sample_geometric_pmf_d ( double p )

Draws a double precision random sample from the Geometric pmf.

Parameters:

p	Probability of success, must be in (0,1).

Returns:: A random sample from the geometric distribution with parameter p.

Warning:: The parameter p must be >= 0.

Note:: Does not seed the rand() function. If this is desired, srand() must be called prior to calling this function.

Definition at line 595 of file pmf.c.

Defines

Functions

Detailed Description

Define Documentation

Function Documentation