1 /* PSPP - a program for statistical analysis.
2 Copyright (C) 2005 Free Software Foundation, Inc.
4 This program is free software: you can redistribute it and/or modify
5 it under the terms of the GNU General Public License as published by
6 the Free Software Foundation, either version 3 of the License, or
7 (at your option) any later version.
9 This program is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 GNU General Public License for more details.
14 You should have received a copy of the GNU General Public License
15 along with this program. If not, see <http://www.gnu.org/licenses/>. */
21 Find the least-squares estimate of b for the linear model:
25 where Y is an n-by-1 column vector, X is an n-by-p matrix of
26 independent variables, b is a p-by-1 vector of regression coefficients,
27 and Z is an n-by-1 normally-distributed random vector with independent
28 identically distributed components with mean 0.
30 This estimate is found via the sweep operator, which is a modification
31 of Gauss-Jordan pivoting.
36 Matrix Computations, third edition. GH Golub and CF Van Loan.
37 The Johns Hopkins University Press. 1996. ISBN 0-8018-5414-8.
39 Numerical Analysis for Statisticians. K Lange. Springer. 1999.
42 Numerical Linear Algebra for Applications in Statistics. JE Gentle.
43 Springer. 1998. ISBN 0-387-98542-5.
48 The matrix A will be overwritten. In ordinary uses of the sweep
49 operator, A will be the matrix
57 X refers to the design matrix and Y to the vector of dependent
58 observations. reg_sweep sweeps on the diagonal elements of
61 The matrix A is assumed to be symmetric, so the sweep operation is
62 performed only for the upper triangle of A.
65 #include <gsl/gsl_matrix.h>
66 #include <gsl/gsl_math.h>
68 int reg_sweep (gsl_matrix * A);