estimate.c File Reference

#include <math.h>
#include <stdlib.h>
#include "tabledesign.h"

Functions
void	acf (double *vec, int n, double *out, int k)
	Computes the autocorrelation of a vector. More...
int	durbin (double *arg0, int n, double *arg2, double *arg3, double *outSomething)
void	afromk (double *in, double *out, int n)
int	kfroma (double *in, double *out, int n)
void	rfroma (double *arg0, int n, double *arg2)
double	model_dist (double *arg0, double *arg1, int n)
void	acmat (short *in, int n, int m, double **out)
void	acvect (short *in, int n, int m, double *out)
int	lud (double **a, int n, int *indx, int *d)
	Replaces a real n-by-n matrix "a" with the LU decomposition of a row-wise permutation of itself. More...
void	lubksb (double **a, int n, int *indx, double *b)
	Solves the set of n linear equations Ax = b, using LU decomposition back-substitution. More...

Function Documentation

acf()

void acf	(	double *	vec,
		int	n,
		double *	out,
		int	k
	)

Computes the autocorrelation of a vector.

More precisely, it computes the dot products of vec[i:] and vec[:-i] for i in [0, k). Unused.

See https://en.wikipedia.org/wiki/Autocorrelation.

acmat()

void acmat	(	short *	in,
		int	n,
		int	m,
		double **	out
	)

acvect()

void acvect	(	short *	in,
		int	n,
		int	m,
		double *	out
	)

afromk()

void afromk	(	double *	in,
		double *	out,
		int	n
	)

durbin()

int durbin	(	double *	arg0,
		int	n,
		double *	arg2,
		double *	arg3,
		double *	outSomething
	)

kfroma()

int kfroma	(	double *	in,
		double *	out,
		int	n
	)

lubksb()

void lubksb	(	double **	a,
		int	n,
		int *	indx,
		double *	b
	)

Solves the set of n linear equations Ax = b, using LU decomposition back-substitution.

Input parameters: a: The LU decomposition of a matrix, created by "lud". n: The size of the matrix. indx: Row permutation vector, created by "lud". b: The vector b in the equation. 1-indexed; is should be of size n+1, and index 0 is not used.

Output parameters: b: The output vector x. 1-indexed.

From "Numerical Recipes in C: The Art of Scientific Computing".

lud()

int lud	(	double **	a,
		int	n,
		int *	indx,
		int *	d
	)

Replaces a real n-by-n matrix "a" with the LU decomposition of a row-wise permutation of itself.

Input parameters: a: The matrix which is operated on. 1-indexed; it should be of size (n+1) x (n+1), and row/column index 0 is not used. n: The size of the matrix.

Output parameters: indx: The row permutation performed. 1-indexed; it should be of size n+1, and index 0 is not used. d: the determinant of the permutation matrix.

Returns 1 to indicate failure if the matrix is singular or has zeroes on the diagonal, 0 on success.

Derived from ludcmp in "Numerical Recipes in C: The Art of Scientific Computing", with modified error handling.

model_dist()

double model_dist	(	double *	arg0,
		double *	arg1,
		int	n
	)

rfroma()

void rfroma	(	double *	arg0,
		int	n,
		double *	arg2
	)