#include <yat/regression/Cox.h>

Public Member Functions
	Cox (void)
	Default Constructor.

	Cox (const Cox &other)
	Copy Constructor. More...

	Cox (Cox &&other)

	~Cox (void)
	Destructor.

Cox &	operator= (const Cox &other)

Cox &	operator= (Cox &&other)
	Move assignment.

void	add (double x, double time, bool event)

void	add (const utility::VectorBase &x, const utility::VectorBase &time, const std::vector< char > &event)
	Add n data points. More...

void	clear (void)

void	train (void)

double	b (void) const

double	z (void) const

double	p (void) const
	two-sided p-value for the null-hypothesis that b is zero.

double	hazard_ratio (void) const

double	hazard_ratio_lower_CI (double alpha=0.95) const

double	hazard_ratio_upper_CI (double alpha=0.95) const

Detailed Description

Class handling univariable Cox regression.

The time-dependent hazard is modeled as $\lambda(t|x) = \lambda_0(t) \exp (bx)$

Model parameters are inferred as described by Breslow (1975) and tied timepoints are handled as suggested by Efron (1975).

Constructor & Destructor Documentation

◆ Cox() [1/2]

theplu::yat::regression::Cox::Cox ( const Cox & other )

Copy Constructor.

Create a copy of other and its data

◆ Cox() [2/2]

theplu::yat::regression::Cox::Cox ( Cox && other )

Move Constructor

Member Function Documentation

◆ add() [1/2]

void theplu::yat::regression::Cox::add	(	double	x,
		double	time,
		bool	event
	)

Add one data point

Parameters

x	value of the covariate variable
time	timepoint when event occured or sample was censored
event	true for non-censored sample

◆ add() [2/2]

void theplu::yat::regression::Cox::add	(	const utility::VectorBase &	x,
		const utility::VectorBase &	time,
		const std::vector< char > &	event
	)

Add n data points.

Equivalent to looping over the vectors and calling single-sample version of add.

◆ b()

double theplu::yat::regression::Cox::b ( void ) const

The time-dependent hazard is modeled as $\lambda(t|x) = \lambda_0(t) \exp (bx)$

◆ clear()

void theplu::yat::regression::Cox::clear ( void )

Remove all data.

◆ hazard_ratio()

double theplu::yat::regression::Cox::hazard_ratio ( void ) const

The hazard ratio is defined as $ exp(b) $ , in other words, the amount the hazard increases when the co-variate x increases with one unit.

◆ hazard_ratio_lower_CI()

double theplu::yat::regression::Cox::hazard_ratio_lower_CI ( double alpha = 0.95 ) const

Returns: the lower bound of the alpha % confidence interval

◆ hazard_ratio_upper_CI()

double theplu::yat::regression::Cox::hazard_ratio_upper_CI ( double alpha = 0.95 ) const

Returns: the upper bound of the alpha % confidence interval

◆ operator=()

Cox& theplu::yat::regression::Cox::operator= ( const Cox & other )

Assignment operator

◆ train()

void theplu::yat::regression::Cox::train ( void )

Infer model parameter b and related, hazard ratio and their confidence intervall.

◆ z()

double theplu::yat::regression::Cox::z ( void ) const

Returns: b divided by standard error of b.

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

yat/regression/Cox.h

Public Member Functions