Internal helper functions. More...

Functions
template<typename Iterator >
std::vector< double >	compute_ranks (Iterator first, Iterator last)
	Helper function to compute ranks.

template<typename Iterator , typename Projection >
std::vector< double >	compute_ranks (Iterator first, Iterator last, Projection proj)
	Rank computation (projection version)

double	link_transform (double mu, link_function link)
	Link function g(mu) -> eta.

double	inverse_link (double eta, link_function link)
	Inverse link function g^{-1}(eta) -> mu.

double	link_derivative (double mu, link_function link)
	Derivative of link function d(eta)/d(mu) = g'(mu)

double	variance_function (double mu, distribution_family family)
	Variance function V(mu)

double	deviance_residual (double y, double mu, distribution_family family)
	Calculate deviance (for a single observation)

std::vector< double >	solve_weighted_least_squares (const std::vector< std::vector< double > > &X, const std::vector< double > &z, const std::vector< double > &w, std::vector< std::vector< double > > &XtWX_inv)
	Solve weighted least squares.

void	validate_matrix_structure (const std::vector< std::vector< double > > &data, const char *func_name)
	Validate 2D matrix structure.

void	validate_no_intercept_column (const std::vector< std::vector< double > > &X, const char *func_name)
	Check that X data does not contain an intercept column.

std::vector< std::vector< double > >	transpose (const std::vector< std::vector< double > > &A)
	Calculate transpose matrix.

std::vector< std::vector< double > >	matrix_multiply (const std::vector< std::vector< double > > &A, const std::vector< std::vector< double > > &B)
	Calculate matrix product.

std::vector< double >	matrix_vector_multiply (const std::vector< std::vector< double > > &A, const std::vector< double > &v)
	Calculate matrix-vector product.

std::vector< std::vector< double > >	cholesky (const std::vector< std::vector< double > > &A)
	Perform Cholesky decomposition.

std::vector< double >	solve_cholesky (const std::vector< std::vector< double > > &L, const std::vector< double > &b)
	Solve system of equations using Cholesky decomposition.

std::vector< std::vector< double > >	inverse_cholesky (const std::vector< std::vector< double > > &L)
	Calculate inverse matrix using Cholesky decomposition.

void	standardize_features (const std::vector< std::vector< double > > &X, std::vector< std::vector< double > > &X_scaled, std::vector< double > &X_mean, std::vector< double > &X_std, std::size_t n, std::size_t p)
	特徴量の標準化 (平均0, 標準偏差1)

std::vector< double >	rescale_coefficients (const std::vector< double > &beta, const std::vector< double > &X_mean, const std::vector< double > &X_std, double y_mean, std::size_t p, bool standardize)
	標準化済み係数を元のスケールに逆変換する

double	noncentrality_parameter_t (double effect_size, double n)
	Calculate noncentrality parameter for t-distribution.

double	noncentrality_parameter_t_two_sample (double effect_size, double n1, double n2)
	Calculate noncentrality parameter for two-sample case.

double	critical_t_two_sided (double alpha, double df)
	Calculate critical value for two-sided t-test.

double	critical_t_one_sided (double alpha, double df)
	Calculate critical value for one-sided t-test.

double	critical_z_two_sided (double alpha)
	Calculate critical value for two-sided normal test.

double	critical_z_one_sided (double alpha)
	Calculate critical value for one-sided normal test.

const char *	alternative_to_string (alternative_hypothesis alt)
	Convert alternative_hypothesis enum to string.

Detailed Description

Internal helper functions.

Function Documentation

◆ alternative_to_string()

const char * statcpp::detail::alternative_to_string ( alternative_hypothesis alt )

inline

Convert alternative_hypothesis enum to string.

Parameters

alt	Alternative hypothesis enum value

Returns: Corresponding string ("two.sided", "greater", or "less")

Definition at line 127 of file power_analysis.hpp.

◆ cholesky()

std::vector< std::vector< double > > statcpp::detail::cholesky ( const std::vector< std::vector< double > > & A )

inline

Perform Cholesky decomposition.

Computes lower triangular matrix L such that L * L^T = A for positive definite symmetric matrix A.

Parameters

A	Positive definite symmetric matrix

Returns: std::vector<std::vector<double>> Lower triangular matrix L

Exceptions

std::runtime_error If matrix is not positive definite

Definition at line 419 of file linear_regression.hpp.

◆ compute_ranks() [1/2]

template<typename Iterator >

std::vector< double > statcpp::detail::compute_ranks	(	Iterator	first,
		Iterator	last
	)

Helper function to compute ranks.

Uses average rank for ties.

Template Parameters

Iterator Iterator type

Parameters

first	Begin iterator
last	End iterator

Returns: Vector of ranks

Definition at line 546 of file correlation_covariance.hpp.

◆ compute_ranks() [2/2]

template<typename Iterator , typename Projection >

std::vector< double > statcpp::detail::compute_ranks	(	Iterator	first,
		Iterator	last,
		Projection	proj
	)

Rank computation (projection version)

Template Parameters

Iterator	Iterator type
Projection	Projection function type

Parameters

first	Begin iterator
last	End iterator
proj	Projection function

Returns: Vector of ranks

Definition at line 596 of file correlation_covariance.hpp.

◆ critical_t_one_sided()

double statcpp::detail::critical_t_one_sided	(	double	alpha,
		double	df
	)

inline

Calculate critical value for one-sided t-test.

Parameters

alpha	Significance level
df	Degrees of freedom

Returns: Critical value

Definition at line 97 of file power_analysis.hpp.

◆ critical_t_two_sided()

double statcpp::detail::critical_t_two_sided	(	double	alpha,
		double	df
	)

inline

Calculate critical value for two-sided t-test.

Parameters

alpha	Significance level
df	Degrees of freedom

Returns: Critical value

Definition at line 86 of file power_analysis.hpp.

◆ critical_z_one_sided()

double statcpp::detail::critical_z_one_sided ( double alpha )

inline

Calculate critical value for one-sided normal test.

Parameters

alpha Significance level

Returns: Critical value

Definition at line 117 of file power_analysis.hpp.

◆ critical_z_two_sided()

double statcpp::detail::critical_z_two_sided ( double alpha )

inline

Calculate critical value for two-sided normal test.

Parameters

alpha Significance level

Returns: Critical value

Definition at line 107 of file power_analysis.hpp.

◆ deviance_residual()

double statcpp::detail::deviance_residual	(	double	y,
		double	mu,
		distribution_family	family
	)

inline

Calculate deviance (for a single observation)

Calculates the deviance residual according to the distribution family.

Parameters

y	Observed value
mu	Expected value
family	Distribution family

Returns: Deviance residual

Definition at line 228 of file glm.hpp.

◆ inverse_cholesky()

std::vector< std::vector< double > > statcpp::detail::inverse_cholesky ( const std::vector< std::vector< double > > & L )

inline

Calculate inverse matrix using Cholesky decomposition.

Parameters

L	Lower triangular matrix from Cholesky decomposition

Returns: std::vector<std::vector<double>> Inverse of the original matrix

Definition at line 489 of file linear_regression.hpp.

◆ inverse_link()

double statcpp::detail::inverse_link	(	double	eta,
		link_function	link
	)

inline

Inverse link function g^{-1}(eta) -> mu.

Transforms the linear predictor eta using the inverse link function to return the expected value mu.

Parameters

eta	Linear predictor
link	Link function to use

Returns: Expected value mu

Definition at line 127 of file glm.hpp.

◆ link_derivative()

double statcpp::detail::link_derivative	(	double	mu,
		link_function	link
	)

inline

Derivative of link function d(eta)/d(mu) = g'(mu)

Calculates the derivative of the link function with respect to the expected value mu.

Parameters

mu	Expected value
link	Link function to use

Returns: Derivative of the link function

Exceptions

std::runtime_error If mu is close to 0 for cloglog link

Definition at line 160 of file glm.hpp.

◆ link_transform()

double statcpp::detail::link_transform	(	double	mu,
		link_function	link
	)

inline

Link function g(mu) -> eta.

Transforms the expected value mu using the link function to return the linear predictor eta.

Parameters

mu	Expected value
link	Link function to use

Returns: Linear predictor eta

Definition at line 95 of file glm.hpp.

◆ matrix_multiply()

std::vector< std::vector< double > > statcpp::detail::matrix_multiply	(	const std::vector< std::vector< double > > &	A,
		const std::vector< std::vector< double > > &	B
	)

inline

Calculate matrix product.

Parameters

A	Left matrix (m x n)
B	Right matrix (n x p)

Returns: std::vector<std::vector<double>> Product matrix (m x p)

Exceptions

std::invalid_argument	If matrix dimensions are incompatible
std::invalid_argument	If matrix rows are inconsistent

Definition at line 345 of file linear_regression.hpp.

◆ matrix_vector_multiply()

std::vector< double > statcpp::detail::matrix_vector_multiply	(	const std::vector< std::vector< double > > &	A,
		const std::vector< double > &	v
	)

inline

Calculate matrix-vector product.

Parameters

A	Matrix (m x n)
v	Vector (n-dimensional)

Returns: std::vector<double> Result vector (m-dimensional)

Exceptions

std::invalid_argument If matrix and vector dimensions are incompatible

Definition at line 389 of file linear_regression.hpp.

◆ noncentrality_parameter_t()

double statcpp::detail::noncentrality_parameter_t	(	double	effect_size,
		double	n
	)

inline

Calculate noncentrality parameter for t-distribution.

Parameters

effect_size	Effect size
n	Sample size

Returns: Noncentrality parameter

Definition at line 63 of file power_analysis.hpp.

◆ noncentrality_parameter_t_two_sample()

double statcpp::detail::noncentrality_parameter_t_two_sample	(	double	effect_size,
		double	n1,
		double	n2
	)

inline

Calculate noncentrality parameter for two-sample case.

Parameters

effect_size	Effect size
n1	Sample size of group 1
n2	Sample size of group 2

Returns: Noncentrality parameter

Definition at line 75 of file power_analysis.hpp.

◆ rescale_coefficients()

std::vector< double > statcpp::detail::rescale_coefficients	(	const std::vector< double > &	beta,
		const std::vector< double > &	X_mean,
		const std::vector< double > &	X_std,
		double	y_mean,
		std::size_t	p,
		bool	standardize
	)

inline

標準化済み係数を元のスケールに逆変換する

Definition at line 424 of file model_selection.hpp.

◆ solve_cholesky()

std::vector< double > statcpp::detail::solve_cholesky	(	const std::vector< std::vector< double > > &	L,
		const std::vector< double > &	b
	)

inline

Solve system of equations using Cholesky decomposition.

Solves A * x = b (where A = L * L^T). Efficiently finds the solution using forward and back substitution.

Parameters

L	Lower triangular matrix from Cholesky decomposition
b	Right-hand side vector

Returns: std::vector<double> Solution vector x

Definition at line 454 of file linear_regression.hpp.

◆ solve_weighted_least_squares()

std::vector< double > statcpp::detail::solve_weighted_least_squares	(	const std::vector< std::vector< double > > &	X,
		const std::vector< double > &	z,
		const std::vector< double > &	w,
		std::vector< std::vector< double > > &	XtWX_inv
	)

inline

Solve weighted least squares.

Computes (X'WX)^{-1} X'Wz using Cholesky decomposition.

Parameters

X	Design matrix
z	Working variable vector
w	Weight vector
XtWX_inv	Output: inverse of (X'WX)

Returns: Weighted least squares solution (coefficient vector)

Exceptions

std::runtime_error If matrix is not positive definite

Definition at line 276 of file glm.hpp.

◆ standardize_features()

void statcpp::detail::standardize_features	(	const std::vector< std::vector< double > > &	X,
		std::vector< std::vector< double > > &	X_scaled,
		std::vector< double > &	X_mean,
		std::vector< double > &	X_std,
		std::size_t	n,
		std::size_t	p
	)

inline

特徴量の標準化 (平均0, 標準偏差1)

Definition at line 393 of file model_selection.hpp.

◆ transpose()

std::vector< std::vector< double > > statcpp::detail::transpose ( const std::vector< std::vector< double > > & A )

inline

Calculate transpose matrix.

Parameters

A	Input matrix

Returns: std::vector<std::vector<double>> Transposed matrix

Exceptions

std::invalid_argument If rows have different column counts

Definition at line 314 of file linear_regression.hpp.

◆ validate_matrix_structure()

void statcpp::detail::validate_matrix_structure	(	const std::vector< std::vector< double > > &	data,
		const char *	func_name
	)

inline

Validate 2D matrix structure.

Verifies that the matrix is not empty and all rows have the same number of columns.

Parameters

data	Matrix to validate
func_name	Function name for error messages

Exceptions

std::invalid_argument	If matrix is empty
std::invalid_argument	If first row is empty
std::invalid_argument	If rows have different column counts

Definition at line 231 of file linear_regression.hpp.

◆ validate_no_intercept_column()

void statcpp::detail::validate_no_intercept_column	(	const std::vector< std::vector< double > > &	X,
		const char *	func_name
	)

inline

Check that X data does not contain an intercept column.

Warns if the first column is all 1.0, as the user may have mistakenly included an intercept column. The intercept is added automatically within the function, so users don't need to include it.

Parameters

X	Predictor matrix
func_name	Function name for error messages

Exceptions

std::invalid_argument If the first column is all 1.0

Definition at line 276 of file linear_regression.hpp.

◆ variance_function()

double statcpp::detail::variance_function	(	double	mu,
		distribution_family	family
	)

inline

Variance function V(mu)

Calculates the variance function according to the distribution family.

Parameters

mu	Expected value
family	Distribution family

Returns: Variance function value

Definition at line 200 of file glm.hpp.

Functions

Detailed Description

Function Documentation

◆ alternative_to_string()

◆ cholesky()

◆ compute_ranks() [1/2]

◆ compute_ranks() [2/2]

◆ critical_t_one_sided()

◆ critical_t_two_sided()

◆ critical_z_one_sided()

◆ critical_z_two_sided()

◆ deviance_residual()

◆ inverse_cholesky()

◆ inverse_link()

◆ link_derivative()

◆ link_transform()

◆ matrix_multiply()

◆ matrix_vector_multiply()

◆ noncentrality_parameter_t()

◆ noncentrality_parameter_t_two_sample()

◆ rescale_coefficients()

◆ solve_cholesky()

◆ solve_weighted_least_squares()

◆ standardize_features()

◆ transpose()

◆ validate_matrix_structure()

◆ validate_no_intercept_column()

◆ variance_function()