npl::StudentsT Class Reference

Student's T-distribution. A cache of the Probability Density Function and cumulative density function is created using the analytical PDF. More...

#include <statistics.h>

Public Member Functions
	StudentsT (int dof=2, double dt=0.1, double tmax=20)
	Defualt constructor takes the degrees of freedom (Nu), step size for numerical CDF computation and Maximum T to sum to for numerical CDF. More...
void	setDOF (double dof)
	Change the degress of freedom, update cache. More...
void	setStepT (double dt)
	Step in t to use for computing the CDF, smaller means more precision although in reality the distribution is quite smooth, and linear interpolation should be very good. More...
void	setMaxT (double tmax)
	Set the maximum t for numerical integration, and recompute the cdf/pdf caches. More...
double	cumulative (double t) const
	Get the cumulative probability at some t value. More...
double	cdf (double t) const
	Get the cumulative probability at some t value. More...
double	density (double t) const
	Get the probability density at some t value. More...
double	pdf (double t) const
	Get the probability density at some t value. More...
double	icdf (double t) const
	Get the T-score that corresponds to a particular p-value. Alias for tthresh. More...
double	tthresh (double p) const
	Get the T-score that corresponds to a particular p-value. Alias for icdf. More...

Detailed Description

Student's T-distribution. A cache of the Probability Density Function and cumulative density function is created using the analytical PDF.

Definition at line 244 of file statistics.h.

Constructor & Destructor Documentation

npl::StudentsT::StudentsT	(	int	dof = 2,
		double	dt = 0.1,
		double	tmax = 20
	)

Defualt constructor takes the degrees of freedom (Nu), step size for numerical CDF computation and Maximum T to sum to for numerical CDF.

Parameters

dof	Degrees of freedom (shape parameter)
dt	step size in x or t to take
tmax	Maximum t to consider

Member Function Documentation

double npl::StudentsT::cdf ( double  t ) const

inline

Get the cumulative probability at some t value.

Parameters

t	T (or x, distance from center) value to query

Returns

Cumulative probability (probability value of value < t)

Definition at line 300 of file statistics.h.

double npl::StudentsT::cumulative

(

double

t

)

const

Get the cumulative probability at some t value.

Parameters

t	T (or x, distance from center) value to query

Returns

Cumulative probability (probability value of value < t)

double npl::StudentsT::density

(

double

t

)

const

Get the probability density at some t value.

Parameters

t	T value to query

Returns

Probability density at t.

double npl::StudentsT::icdf

(

double

t

)

const

Get the T-score that corresponds to a particular p-value. Alias for tthresh.

Parameters

t	t value to get the inverse CDF (area under the curve) for

Returns

T value that matches p.

double npl::StudentsT::pdf ( double  t ) const

inline

Get the probability density at some t value.

Parameters

t	T value to query

Returns

Probability density at t.

Definition at line 318 of file statistics.h.

void npl::StudentsT::setDOF

(

double

dof

)

Change the degress of freedom, update cache.

Parameters

dof	Shape parameter, higher values make the distribution more gaussian

void npl::StudentsT::setMaxT

(

double

tmax

)

Set the maximum t for numerical integration, and recompute the cdf/pdf caches.

Parameters

tmax	CDF and PDF are stored as arrays, this is the maximum acceptable t value. Its RARE (like 10^-10 rare) to have a value higher than 20.

void npl::StudentsT::setStepT

(

double

dt

)

Step in t to use for computing the CDF, smaller means more precision although in reality the distribution is quite smooth, and linear interpolation should be very good.

Parameters

dt	Step size for numerical integration

double npl::StudentsT::tthresh ( double  p ) const

inline

Get the T-score that corresponds to a particular p-value. Alias for icdf.

Parameters

p	probability value is less than this

Returns

T value that matches p.

Definition at line 338 of file statistics.h.

---

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

• /home/micahc/coding/npl/lib/statistics.h