adf< LAG_policy > Class Template Reference

Template class for performing pointwise ADF for a fts. More...

#include <ADF_test.hpp>

Public Member Functions
template<typename STOR_OBJ>
	adf (STOR_OBJ &&x, int k)
	Constructor taking the matrix containing the fts and the lag order.
std::vector< double >	p_values () const
	Getter for the poitwise ADF test p-values.
std::vector< double >	one_step_diff (const KO_Traits::StoringVector &ts) const
	Computing the one time step difference for the time series.
KO_Traits::StoringMatrix	embed (const KO_Traits::StoringVector &ts, int dimension) const
	Embeds the time series into a low-dimensional Euclidean space.
double	statistic_eval (const KO_Traits::StoringVector &ts) const
	Evaluating the ADF-test statistic for a time series.
double	p_value_eval (const KO_Traits::StoringVector &ts, const std::vector< double > &tableipl, int i) const
	Evaluating thr ADF-test p-value for a time series.
void	test ()
	computing the ADF-test pointwisely for each point of the domain for the fts

Detailed Description

template<class LAG_policy>
class adf< LAG_policy >

Template class for performing pointwise ADF for a fts.

Template Parameters

LAG_policy

indicates if lag orders bigger than one has to be taken into account while computing the test statistic

The ADF is performed for all the rows of the matrix storing the fts as: every row indicates a domain point for which an evaluation of the functional object is available, every column a time instant.

Constructor & Destructor Documentation

adf()

template<class LAG_policy>

template<typename STOR_OBJ>

adf< LAG_policy >::adf ( STOR_OBJ && x, int k )

inline

Constructor taking the matrix containing the fts and the lag order.

Parameters

x	matrix containing the fts
k	lag order

Universal constructor: move semantic used to optimazing handling big size objects

Member Function Documentation

embed()

template<class LAG_policy>

KO_Traits::StoringMatrix adf< LAG_policy >::embed	(	const KO_Traits::StoringVector &	ts,
		int	dimension ) const

Embeds the time series into a low-dimensional Euclidean space.

Parameters

ts	time series
dimension	dimension of the Euclidean space in which embedding the time series

Returns

a matrix containing the embedded time series

Each row of the resulting matrix consists of sequences x[t], x[t-1], …, x[t-dimension+1], where t is the original index of x.

Note

Implementation taken from (https://rdocumentation.org/packages/stats/versions/3.6.2)

one_step_diff()

template<class LAG_policy>

std::vector< double > adf< LAG_policy >::one_step_diff

(

const KO_Traits::StoringVector &

ts

)

const

Computing the one time step difference for the time series.

Parameters

ts	the time series

Returns

an std::vector<double> containing the one time step differences of the time series

p_value_eval()

template<class LAG_policy>

double adf< LAG_policy >::p_value_eval	(	const KO_Traits::StoringVector &	ts,
		const std::vector< double > &	tableipl,
		int	i ) const

Evaluating thr ADF-test p-value for a time series.

Parameters

ts	time series
tableipl	table containing extreme values for the statistic for p-value computation
i	number of the discrete evaluation corresponding to 'ts'

Returns

ADF-test p-value

p_values()

template<class LAG_policy>

std::vector< double > adf< LAG_policy >::p_values ( ) const

inline

Getter for the poitwise ADF test p-values.

Returns

the private m_p_values

statistic_eval()

template<class LAG_policy>

double adf< LAG_policy >::statistic_eval

(

const KO_Traits::StoringVector &

ts

)

const

Evaluating the ADF-test statistic for a time series.

Parameters

ts	time series

Returns

statistic evaluation

---

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

• src/ADF_test.hpp
• src/ADF_test_imp.hpp