stats< T > Class Template Reference

Collaboration diagram for stats< T >:

Public Member Functions
	stats ()
	Constructor.
	~stats ()
	Destructor.
T	mean (std::vector< T > &X)
	The populational mean.
T	findS (std::vector< T > &X, T Xbar)
	The standard deviation of the population.
T	findS (std::vector< T > &X, T Xbar, T adj)
	The standard deviation of the population, adjust for negative numbers.
T	rRegression (std::vector< T > &X, std::vector< T > &Y)
	basic pearson r linear regression
T	bRegression (std::vector< T > &X, std::vector< T > &Y)
	regression coefficent slope
T	Bxy (T r, T Sx, T Sy)
	raw score regression coefficent
T	simpleANOVA (std::vector< T > &X, std::vector< T > &Y)
	This is a simple ANOVA for two groups.
T	decisionGGC (T mu1, T mu2, T sigma1, T sigma2, T PofA)
	find the decision boundry as described in Itti(2000) PhD Thesis pg. 145-8
T	getDPrime ()
	Return D' after running descisionGGC.
T	getErrorGGC_2AFC (T mu1, T mu2, T sigma1, T sigma2)
	Get the probability of Error for a 2AFC paradigm.
T	gauss (T x, T mu, T sigma)
	return the gaussian from f(x;mu,sigma) = guassian
Public Attributes
bool	GGC
	some bools for ASSERTION testing
T	Xb
	X bar - the population mean for x.
T	Yb
	Y bar - the population mean for y.
T	S2
	S squared - population measure for variance.
T	Sx
	S - population measure for varaince.
T	Sy
T	S
T	r
	r - the corralation coefficent
T	b
	b - the regression coefficent slope
T	PA
	probability of A or B
T	PB
T	D
	decision boundrys
T	Dprime
T	SStotal
	SStotal, SSwithin, SSbetween.
T	SSwithin
T	SSbetween
T	DFwithin
T	DFbetween
T	MSwithin
T	MSbetween
T	F

Detailed Description

template<class T>
class stats< T >

Definition at line 54 of file stats.H.

Constructor & Destructor Documentation

template<class T >

stats< T >::stats ( ) [inline]

Constructor.

Definition at line 53 of file stats.C.

References stats< T >::GGC.

template<class T >

stats< T >::~stats ( ) [inline]

Destructor.

Definition at line 58 of file stats.C.

Member Function Documentation

template<class T>

T stats< T >::bRegression	(	std::vector< T > &	X,
		std::vector< T > &	Y
	)

regression coefficent slope

This can be use to fit the line of regression as Y' = b*(X - Xbar) + YBar

template<class T>

T stats< T >::Bxy	(	T	r,
		T	Sx,
		T	Sy
	)			`[inline]`

raw score regression coefficent

This can be use to fit the line of regression as Y' = Bxy*(X - Xbar) + YBar

Definition at line 145 of file stats.C.

template<class T>

T stats< T >::decisionGGC	(	T	mu1,
		T	mu2,
		T	sigma1,
		T	sigma2,
		T	PofA
	)

find the decision boundry as described in Itti(2000) PhD Thesis pg. 145-8

Find a decision in the general gaussian case input mu's sigma's and P(X)'s for two events. This will return D. Use getDPrime() to get D' following this command.

Parameters:

	mu1	mean for condition 1
	mu2	mean for condition 2
	sigma1	std dev for condition 1
	sigma2	std dev for condition 2
	PofA	probability of A (0 to 1) PofB is determined as 1 - PofA

template<class T>

T stats< T >::findS	(	std::vector< T > &	X,
		T	Xbar,
		T	adj
	)			`[inline]`

The standard deviation of the population, adjust for negative numbers.

Definition at line 93 of file stats.C.

References ASSERT, stats< T >::S2, and sqrt().

template<class T>

T stats< T >::findS	(	std::vector< T > &	X,
		T	Xbar
	)			`[inline]`

The standard deviation of the population.

Definition at line 77 of file stats.C.

References ASSERT, stats< T >::S2, and sqrt().

Referenced by EyeTrackerISCAN::getCalibrationSet(), stats< T >::rRegression(), CINNICstatsRun::runPointAndFlood(), and CINNICstatsRun::runStandardStats().

template<class T>

T stats< T >::gauss	(	T	x,
		T	mu,
		T	sigma
	)			`[inline]`

return the gaussian from f(x;mu,sigma) = guassian

Return a simple P(x) based upon the gaussian distribution with an input sample x and gaussian defined with mu as E(x) and sigma as E(x^2)

Definition at line 229 of file stats.C.

References exp(), and sqrt().

Referenced by segmentImageMerge2::SIMverganceSpring(), and segmentImageMerge::verganceSpring().

template<class T >

T stats< T >::getDPrime ( ) [inline]

Return D' after running descisionGGC.

Definition at line 213 of file stats.C.

References ASSERT, and stats< T >::GGC.

template<class T>

T stats< T >::getErrorGGC_2AFC	(	T	mu1,
		T	mu2,
		T	sigma1,
		T	sigma2
	)			`[inline]`

Get the probability of Error for a 2AFC paradigm.

take the decision boundrys and find the probability of error from them. This is for Two Alternative forced choice paradigm

Parameters:

	mu1	mean for condition 1
	mu2	mean for condition 2
	sigma1	std dev for condition 1
	sigma2	std dev for condition 2
	PofA	probability of A (0 to 1) PofB is determined as 1 - PofA