X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fmath%2Flinreg%2Flinreg.h;h=a7d408af56bde0c605c648963882545d9bec2967;hb=b5b474193e450bba97610065df0518c08074a7fb;hp=2852dd77829ad69f6b53aa993c4386dfd57c1bea;hpb=f4ac26fb5880f9aeba2c7303d0e687b183bc3da1;p=pspp-builds.git
diff --git a/src/math/linreg/linreg.h b/src/math/linreg/linreg.h
index 2852dd77..a7d408af 100644
--- a/src/math/linreg/linreg.h
+++ b/src/math/linreg/linreg.h
@@ -1,39 +1,35 @@
-/* lib/linreg/linreg.h
+/* PSPP - a program for statistical analysis.
+ Copyright (C) 2005 Free Software Foundation, Inc. Written by Jason H. Stover.
- Copyright (C) 2005 Free Software Foundation, Inc.
- Written by Jason H. Stover.
+ This program is free software: you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation, either version 3 of the License, or
+ (at your option) any later version.
- This program is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 2 of the License, or (at
- your option) any later version.
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
- This program is distributed in the hope that it will be useful, but
- WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- General Public License for more details.
-
- You should have received a copy of the GNU General Public License
- along with this program; if not, write to the Free Software
- Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
- 02111-1307, USA.
-*/
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see . */
#ifndef LINREG_H
#define LINREG_H
-
-
+#include
+#include
#include
#include
-struct variable ;
-struct pspp_linreg_coeff;
+struct variable;
+struct pspp_coeff;
union value;
enum
{
+ PSPP_LINREG_CONDITIONAL_INVERSE,
+ PSPP_LINREG_QR,
PSPP_LINREG_SWEEP,
- PSPP_LINREG_SVD
};
@@ -43,23 +39,27 @@ enum
*/
struct pspp_linreg_opts_struct
{
- int resid; /* Should the residuals be returned? */
-
int get_depvar_mean_std;
- int *get_indep_mean_std; /* Array of booleans dictating which
- independent variables need their means
- and standard deviations computed within
- pspp_linreg. This array MUST be of
- length n_indeps. If element i is 1,
- pspp_linreg will compute the mean and
- variance of indpendent variable i. If
- element i is 0, it will not compute the
- mean and standard deviation, and assume
- the values are stored.
- cache->indep_mean[i] is the mean and
- cache->indep_std[i] is the sample
- standard deviation.
- */
+ int *get_indep_mean_std; /* Array of booleans
+ dictating which
+ independent variables need
+ their means and standard
+ deviations computed within
+ pspp_linreg. This array
+ MUST be of length
+ n_indeps. If element i is
+ 1, pspp_linreg will
+ compute the mean and
+ variance of indpendent
+ variable i. If element i
+ is 0, it will not compute
+ the mean and standard
+ deviation, and assume the
+ values are stored.
+ cache->indep_mean[i] is
+ the mean and
+ cache->indep_std[i] is the
+ sample standard deviation. */
};
typedef struct pspp_linreg_opts_struct pspp_linreg_opts;
@@ -69,7 +69,7 @@ typedef struct pspp_linreg_opts_struct pspp_linreg_opts;
Y = Xb + Z
- where Y is an n-by-1 column vector, X is an n-by-p matrix of
+ where Y is an n-by-1 column vector, X is an n-by-p matrix of
independent variables, b is a p-by-1 vector of regression coefficients,
and Z is an n-by-1 normally-distributed random vector with independent
identically distributed components with mean 0.
@@ -95,17 +95,18 @@ struct pspp_linreg_cache_struct
{
int n_obs; /* Number of observations. */
int n_indeps; /* Number of independent variables. */
- int n_coeffs;
+ int n_coeffs; /* The intercept is not considered a
+ coefficient here. */
- /*
- The variable struct is ignored during estimation.
- It is here so the calling procedure can
- find the variable used in the model.
- */
+ /*
+ The variable struct is ignored during estimation. It is here so
+ the calling procedure can find the variable used in the model.
+ */
const struct variable *depvar;
gsl_vector *residuals;
- struct pspp_linreg_coeff *coeff;
+ struct pspp_coeff **coeff;
+ double intercept;
int method; /* Method to use to estimate parameters. */
/*
Means and standard deviations of the variables.
@@ -125,19 +126,19 @@ struct pspp_linreg_cache_struct
Sums of squares.
*/
double ssm; /* Sums of squares for the overall model. */
- gsl_vector *ss_indeps; /* Sums of squares from each
- independent variable.
- */
+ gsl_vector *ss_indeps; /* Sums of squares from each
+ independent variable. */
double sst; /* Sum of squares total. */
double sse; /* Sum of squares error. */
- double mse; /* Mean squared error. This is just sse / dfe, but
- since it is the best unbiased estimate of the population
- variance, it has its own entry here.
+ double mse; /* Mean squared error. This is just sse /
+ dfe, but since it is the best unbiased
+ estimate of the population variance, it
+ has its own entry here. */
+ gsl_vector *ssx; /* Centered sums of squares for independent
+ variables, i.e. \sum (x[i] - mean(x))^2. */
+ double ssy; /* Centered sums of squares for dependent
+ variable.
*/
- gsl_vector *ssx; /* Centered sums of squares for independent variables,
- i.e. \sum (x[i] - mean(x))^2.
- */
- double ssy; /* Centered sums of squares for dependent variable. */
/*
Covariance matrix of the parameter estimates.
*/
@@ -154,6 +155,19 @@ struct pspp_linreg_cache_struct
design matrix.
*/
gsl_matrix *hat;
+
+ double (*predict) (const struct variable **, const union value **,
+ const void *, int);
+ double (*residual) (const struct variable **,
+ const union value **,
+ const union value *, const void *, int);
+ /*
+ Returns pointers to the variables used in the model.
+ */
+ int (*get_vars) (const void *, const struct variable **);
+ struct variable *resid;
+ struct variable *pred;
+
};
typedef struct pspp_linreg_cache_struct pspp_linreg_cache;
@@ -162,23 +176,30 @@ typedef struct pspp_linreg_cache_struct pspp_linreg_cache;
/*
Allocate a pspp_linreg_cache and return a pointer
- to it. n is the number of cases, p is the number of
+ to it. n is the number of cases, p is the number of
independent variables.
*/
-pspp_linreg_cache * pspp_linreg_cache_alloc (size_t n, size_t p);
+pspp_linreg_cache *pspp_linreg_cache_alloc (size_t n, size_t p);
-void pspp_linreg_cache_free (pspp_linreg_cache * c);
+bool pspp_linreg_cache_free (void *);
/*
Fit the linear model via least squares. All pointers passed to pspp_linreg
are assumed to be allocated to the correct size and initialized to the
- values as indicated by opts.
+ values as indicated by opts.
*/
-int pspp_linreg (const gsl_vector * Y, const gsl_matrix * X,
- const pspp_linreg_opts * opts,
- pspp_linreg_cache * cache);
+int
+pspp_linreg (const gsl_vector * Y, const gsl_matrix * X,
+ const pspp_linreg_opts * opts, pspp_linreg_cache * cache);
double
-pspp_linreg_predict (const struct variable *, const union value *,
- const pspp_linreg_cache *, int);
+pspp_linreg_predict (const struct variable **, const union value **,
+ const void *, int);
+double
+pspp_linreg_residual (const struct variable **, const union value **,
+ const union value *, const void *, int);
+/*
+ All variables used in the model.
+ */
+int pspp_linreg_get_vars (const void *, const struct variable **);
#endif