/* PSPP - a program for statistical analysis.
- Copyright (C) 2005 Free Software Foundation, Inc.
+ Copyright (C) 2005, 2009, 2010 Free Software Foundation, Inc.
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
#include <gsl/gsl_vector.h>
#include <math.h>
#include <stdlib.h>
-
-#include "regression-export.h"
#include <data/case.h>
#include <data/casegrouper.h>
#include <data/casereader.h>
-#include <data/category.h>
#include <data/dictionary.h>
#include <data/missing-values.h>
#include <data/procedure.h>
#include <language/dictionary/split-file.h>
#include <language/data-io/file-handle.h>
#include <language/lexer/lexer.h>
-#include <libpspp/alloc.h>
#include <libpspp/compiler.h>
#include <libpspp/message.h>
#include <libpspp/taint.h>
-#include <math/design-matrix.h>
-#include <math/coefficient.h>
-#include <math/linreg/linreg.h>
+#include <math/covariance.h>
+#include <math/linreg.h>
#include <math/moments.h>
-#include <output/table.h>
+#include <output/tab.h>
+
+#include "xalloc.h"
#include "gettext.h"
#define _(msgid) gettext (msgid)
f,
defaults,
all;
- export=custom;
^dependent=varlist;
+save[sv_]=resid,pred;
+method=enter.
const struct variable *v;
};
-/* Linear regression models. */
-static pspp_linreg_cache **models = NULL;
-
/*
Transformations for saving predicted values
and residuals, etc.
{
int n_trns; /* Number of transformations. */
int trns_id; /* Which trns is this one? */
- pspp_linreg_cache *c; /* Linear model for this trns. */
+ linreg *c; /* Linear model for this trns. */
};
/*
Variables used (both explanatory and response).
*/
static size_t n_variables;
-/*
- File where the model will be saved if the EXPORT subcommand
- is given.
- */
-static struct file_handle *model_file;
-
static bool run_regression (struct casereader *, struct cmd_regression *,
- struct dataset *);
+ struct dataset *, linreg **);
/*
STATISTICS subcommand output functions.
*/
-static void reg_stats_r (pspp_linreg_cache *);
-static void reg_stats_coeff (pspp_linreg_cache *);
-static void reg_stats_anova (pspp_linreg_cache *);
-static void reg_stats_outs (pspp_linreg_cache *);
-static void reg_stats_zpp (pspp_linreg_cache *);
-static void reg_stats_label (pspp_linreg_cache *);
-static void reg_stats_sha (pspp_linreg_cache *);
-static void reg_stats_ci (pspp_linreg_cache *);
-static void reg_stats_f (pspp_linreg_cache *);
-static void reg_stats_bcov (pspp_linreg_cache *);
-static void reg_stats_ses (pspp_linreg_cache *);
-static void reg_stats_xtx (pspp_linreg_cache *);
-static void reg_stats_collin (pspp_linreg_cache *);
-static void reg_stats_tol (pspp_linreg_cache *);
-static void reg_stats_selection (pspp_linreg_cache *);
-static void statistics_keyword_output (void (*)(pspp_linreg_cache *),
- int, pspp_linreg_cache *);
+static void reg_stats_r (linreg *, void *);
+static void reg_stats_coeff (linreg *, void *);
+static void reg_stats_anova (linreg *, void *);
+static void reg_stats_outs (linreg *, void *);
+static void reg_stats_zpp (linreg *, void *);
+static void reg_stats_label (linreg *, void *);
+static void reg_stats_sha (linreg *, void *);
+static void reg_stats_ci (linreg *, void *);
+static void reg_stats_f (linreg *, void *);
+static void reg_stats_bcov (linreg *, void *);
+static void reg_stats_ses (linreg *, void *);
+static void reg_stats_xtx (linreg *, void *);
+static void reg_stats_collin (linreg *, void *);
+static void reg_stats_tol (linreg *, void *);
+static void reg_stats_selection (linreg *, void *);
+static void statistics_keyword_output (void (*)(linreg *, void *),
+ int, linreg *, void *);
static void
-reg_stats_r (pspp_linreg_cache * c)
+reg_stats_r (linreg *c, void *aux UNUSED)
{
struct tab_table *t;
int n_rows = 2;
double std_error;
assert (c != NULL);
- rsq = c->ssm / c->sst;
- adjrsq = 1.0 - (1.0 - rsq) * (c->n_obs - 1.0) / (c->n_obs - c->n_indeps);
- std_error = sqrt ((c->n_indeps - 1.0) / (c->n_obs - 1.0));
- t = tab_create (n_cols, n_rows, 0);
- tab_dim (t, tab_natural_dimensions);
+ rsq = linreg_ssreg (c) / linreg_sst (c);
+ adjrsq = 1.0 - (1.0 - rsq) * (linreg_n_obs (c) - 1.0) / (linreg_n_obs (c) - linreg_n_coeffs (c));
+ std_error = sqrt (linreg_mse (c));
+ t = tab_create (n_cols, n_rows);
tab_box (t, TAL_2, TAL_2, -1, TAL_1, 0, 0, n_cols - 1, n_rows - 1);
tab_hline (t, TAL_2, 0, n_cols - 1, 1);
tab_vline (t, TAL_2, 2, 0, n_rows - 1);
tab_text (t, 2, 0, TAB_CENTER | TAT_TITLE, _("R Square"));
tab_text (t, 3, 0, TAB_CENTER | TAT_TITLE, _("Adjusted R Square"));
tab_text (t, 4, 0, TAB_CENTER | TAT_TITLE, _("Std. Error of the Estimate"));
- tab_float (t, 1, 1, TAB_RIGHT, sqrt (rsq), 10, 2);
- tab_float (t, 2, 1, TAB_RIGHT, rsq, 10, 2);
- tab_float (t, 3, 1, TAB_RIGHT, adjrsq, 10, 2);
- tab_float (t, 4, 1, TAB_RIGHT, std_error, 10, 2);
+ tab_double (t, 1, 1, TAB_RIGHT, sqrt (rsq), NULL);
+ tab_double (t, 2, 1, TAB_RIGHT, rsq, NULL);
+ tab_double (t, 3, 1, TAB_RIGHT, adjrsq, NULL);
+ tab_double (t, 4, 1, TAB_RIGHT, std_error, NULL);
tab_title (t, _("Model Summary"));
tab_submit (t);
}
Table showing estimated regression coefficients.
*/
static void
-reg_stats_coeff (pspp_linreg_cache * c)
+reg_stats_coeff (linreg * c, void *aux_)
{
size_t j;
int n_cols = 7;
int n_rows;
+ int this_row;
double t_stat;
double pval;
- double coeff;
double std_err;
double beta;
const char *label;
- char *tmp;
+
const struct variable *v;
- const union value *val;
- const char *val_s;
struct tab_table *t;
+ gsl_matrix *cov = aux_;
assert (c != NULL);
- tmp = xnmalloc (MAX_STRING, sizeof (*tmp));
- n_rows = c->n_coeffs + 2;
+ n_rows = linreg_n_coeffs (c) + 3;
- t = tab_create (n_cols, n_rows, 0);
+ t = tab_create (n_cols, n_rows);
tab_headers (t, 2, 0, 1, 0);
- tab_dim (t, tab_natural_dimensions);
tab_box (t, TAL_2, TAL_2, -1, TAL_1, 0, 0, n_cols - 1, n_rows - 1);
tab_hline (t, TAL_2, 0, n_cols - 1, 1);
tab_vline (t, TAL_2, 2, 0, n_rows - 1);
tab_text (t, 5, 0, TAB_CENTER | TAT_TITLE, _("t"));
tab_text (t, 6, 0, TAB_CENTER | TAT_TITLE, _("Significance"));
tab_text (t, 1, 1, TAB_LEFT | TAT_TITLE, _("(Constant)"));
- coeff = c->coeff[0]->estimate;
- tab_float (t, 2, 1, 0, coeff, 10, 2);
- std_err = sqrt (gsl_matrix_get (c->cov, 0, 0));
- tab_float (t, 3, 1, 0, std_err, 10, 2);
- beta = coeff / c->depvar_std;
- tab_float (t, 4, 1, 0, beta, 10, 2);
- t_stat = coeff / std_err;
- tab_float (t, 5, 1, 0, t_stat, 10, 2);
- pval = 2 * gsl_cdf_tdist_Q (fabs (t_stat), 1.0);
- tab_float (t, 6, 1, 0, pval, 10, 2);
- for (j = 1; j <= c->n_indeps; j++)
+ tab_double (t, 2, 1, 0, linreg_intercept (c), NULL);
+ std_err = sqrt (gsl_matrix_get (linreg_cov (c), 0, 0));
+ tab_double (t, 3, 1, 0, std_err, NULL);
+ tab_double (t, 4, 1, 0, 0.0, NULL);
+ t_stat = linreg_intercept (c) / std_err;
+ tab_double (t, 5, 1, 0, t_stat, NULL);
+ pval = 2 * gsl_cdf_tdist_Q (fabs (t_stat), (double) (linreg_n_obs (c) - linreg_n_coeffs (c)));
+ tab_double (t, 6, 1, 0, pval, NULL);
+ for (j = 0; j < linreg_n_coeffs (c); j++)
{
- v = pspp_coeff_get_var (c->coeff[j], 0);
+ struct string tstr;
+ ds_init_empty (&tstr);
+ this_row = j + 2;
+
+ v = linreg_indep_var (c, j);
label = var_to_string (v);
/* Do not overwrite the variable's name. */
- strncpy (tmp, label, MAX_STRING);
- if (var_is_alpha (v))
- {
- /*
- Append the value associated with this coefficient.
- This makes sense only if we us the usual binary encoding
- for that value.
- */
-
- val = pspp_coeff_get_value (c->coeff[j], v);
- val_s = var_get_value_name (v, val);
- strncat (tmp, val_s, MAX_STRING);
- }
-
- tab_text (t, 1, j + 1, TAB_CENTER, tmp);
+ ds_put_cstr (&tstr, label);
+ tab_text (t, 1, this_row, TAB_CENTER, ds_cstr (&tstr));
/*
Regression coefficients.
*/
- coeff = c->coeff[j]->estimate;
- tab_float (t, 2, j + 1, 0, coeff, 10, 2);
+ tab_double (t, 2, this_row, 0, linreg_coeff (c, j), NULL);
/*
Standard error of the coefficients.
*/
- std_err = sqrt (gsl_matrix_get (c->cov, j, j));
- tab_float (t, 3, j + 1, 0, std_err, 10, 2);
+ std_err = sqrt (gsl_matrix_get (linreg_cov (c), j + 1, j + 1));
+ tab_double (t, 3, this_row, 0, std_err, NULL);
/*
- 'Standardized' coefficient, i.e., regression coefficient
+ Standardized coefficient, i.e., regression coefficient
if all variables had unit variance.
*/
- beta = gsl_vector_get (c->indep_std, j);
- beta *= coeff / c->depvar_std;
- tab_float (t, 4, j + 1, 0, beta, 10, 2);
+ beta = sqrt (gsl_matrix_get (cov, j, j));
+ beta *= linreg_coeff (c, j) /
+ sqrt (gsl_matrix_get (cov, cov->size1 - 1, cov->size2 - 1));
+ tab_double (t, 4, this_row, 0, beta, NULL);
/*
Test statistic for H0: coefficient is 0.
*/
- t_stat = coeff / std_err;
- tab_float (t, 5, j + 1, 0, t_stat, 10, 2);
+ t_stat = linreg_coeff (c, j) / std_err;
+ tab_double (t, 5, this_row, 0, t_stat, NULL);
/*
P values for the test statistic above.
*/
pval =
2 * gsl_cdf_tdist_Q (fabs (t_stat),
- (double) (c->n_obs - c->n_coeffs));
- tab_float (t, 6, j + 1, 0, pval, 10, 2);
+ (double) (linreg_n_obs (c) - linreg_n_coeffs (c)));
+ tab_double (t, 6, this_row, 0, pval, NULL);
+ ds_destroy (&tstr);
}
tab_title (t, _("Coefficients"));
tab_submit (t);
- free (tmp);
}
/*
Display the ANOVA table.
*/
static void
-reg_stats_anova (pspp_linreg_cache * c)
+reg_stats_anova (linreg * c, void *aux UNUSED)
{
int n_cols = 7;
int n_rows = 4;
- const double msm = c->ssm / c->dfm;
- const double mse = c->sse / c->dfe;
+ const double msm = linreg_ssreg (c) / linreg_dfmodel (c);
+ const double mse = linreg_mse (c);
const double F = msm / mse;
const double pval = gsl_cdf_fdist_Q (F, c->dfm, c->dfe);
struct tab_table *t;
assert (c != NULL);
- t = tab_create (n_cols, n_rows, 0);
+ t = tab_create (n_cols, n_rows);
tab_headers (t, 2, 0, 1, 0);
- tab_dim (t, tab_natural_dimensions);
tab_box (t, TAL_2, TAL_2, -1, TAL_1, 0, 0, n_cols - 1, n_rows - 1);
tab_text (t, 1, 3, TAB_LEFT | TAT_TITLE, _("Total"));
/* Sums of Squares */
- tab_float (t, 2, 1, 0, c->ssm, 10, 2);
- tab_float (t, 2, 3, 0, c->sst, 10, 2);
- tab_float (t, 2, 2, 0, c->sse, 10, 2);
+ tab_double (t, 2, 1, 0, linreg_ssreg (c), NULL);
+ tab_double (t, 2, 3, 0, linreg_sst (c), NULL);
+ tab_double (t, 2, 2, 0, linreg_sse (c), NULL);
/* Degrees of freedom */
- tab_float (t, 3, 1, 0, c->dfm, 4, 0);
- tab_float (t, 3, 2, 0, c->dfe, 4, 0);
- tab_float (t, 3, 3, 0, c->dft, 4, 0);
+ tab_text_format (t, 3, 1, TAB_RIGHT, "%g", c->dfm);
+ tab_text_format (t, 3, 2, TAB_RIGHT, "%g", c->dfe);
+ tab_text_format (t, 3, 3, TAB_RIGHT, "%g", c->dft);
/* Mean Squares */
+ tab_double (t, 4, 1, TAB_RIGHT, msm, NULL);
+ tab_double (t, 4, 2, TAB_RIGHT, mse, NULL);
- tab_float (t, 4, 1, TAB_RIGHT, msm, 8, 3);
- tab_float (t, 4, 2, TAB_RIGHT, mse, 8, 3);
-
- tab_float (t, 5, 1, 0, F, 8, 3);
+ tab_double (t, 5, 1, 0, F, NULL);
- tab_float (t, 6, 1, 0, pval, 8, 3);
+ tab_double (t, 6, 1, 0, pval, NULL);
tab_title (t, _("ANOVA"));
tab_submit (t);
}
+
static void
-reg_stats_outs (pspp_linreg_cache * c)
+reg_stats_outs (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
+
static void
-reg_stats_zpp (pspp_linreg_cache * c)
+reg_stats_zpp (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
+
static void
-reg_stats_label (pspp_linreg_cache * c)
+reg_stats_label (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
+
static void
-reg_stats_sha (pspp_linreg_cache * c)
+reg_stats_sha (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
static void
-reg_stats_ci (pspp_linreg_cache * c)
+reg_stats_ci (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
static void
-reg_stats_f (pspp_linreg_cache * c)
+reg_stats_f (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
static void
-reg_stats_bcov (pspp_linreg_cache * c)
+reg_stats_bcov (linreg * c, void *aux UNUSED)
{
int n_cols;
int n_rows;
assert (c != NULL);
n_cols = c->n_indeps + 1 + 2;
n_rows = 2 * (c->n_indeps + 1);
- t = tab_create (n_cols, n_rows, 0);
+ t = tab_create (n_cols, n_rows);
tab_headers (t, 2, 0, 1, 0);
- tab_dim (t, tab_natural_dimensions);
tab_box (t, TAL_2, TAL_2, -1, TAL_1, 0, 0, n_cols - 1, n_rows - 1);
tab_hline (t, TAL_2, 0, n_cols - 1, 1);
tab_vline (t, TAL_2, 2, 0, n_rows - 1);
tab_vline (t, TAL_0, 1, 0, 0);
tab_text (t, 0, 0, TAB_CENTER | TAT_TITLE, _("Model"));
tab_text (t, 1, 1, TAB_CENTER | TAT_TITLE, _("Covariances"));
- for (i = 1; i < c->n_coeffs; i++)
+ for (i = 0; i < linreg_n_coeffs (c); i++)
{
- const struct variable *v = pspp_coeff_get_var (c->coeff[i], 0);
+ const struct variable *v = linreg_indep_var (c, i);
label = var_to_string (v);
tab_text (t, 2, i, TAB_CENTER, label);
tab_text (t, i + 2, 0, TAB_CENTER, label);
- for (k = 1; k < c->n_coeffs; k++)
+ for (k = 1; k < linreg_n_coeffs (c); k++)
{
col = (i <= k) ? k : i;
row = (i <= k) ? i : k;
- tab_float (t, k + 2, i, TAB_CENTER,
- gsl_matrix_get (c->cov, row, col), 8, 3);
+ tab_double (t, k + 2, i, TAB_CENTER,
+ gsl_matrix_get (c->cov, row, col), NULL);
}
}
tab_title (t, _("Coefficient Correlations"));
tab_submit (t);
}
static void
-reg_stats_ses (pspp_linreg_cache * c)
+reg_stats_ses (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
static void
-reg_stats_xtx (pspp_linreg_cache * c)
+reg_stats_xtx (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
static void
-reg_stats_collin (pspp_linreg_cache * c)
+reg_stats_collin (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
static void
-reg_stats_tol (pspp_linreg_cache * c)
+reg_stats_tol (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
static void
-reg_stats_selection (pspp_linreg_cache * c)
+reg_stats_selection (linreg * c, void *aux UNUSED)
{
assert (c != NULL);
}
static void
-statistics_keyword_output (void (*function) (pspp_linreg_cache *),
- int keyword, pspp_linreg_cache * c)
+statistics_keyword_output (void (*function) (linreg *, void *),
+ int keyword, linreg * c, void *aux)
{
if (keyword)
{
- (*function) (c);
+ (*function) (c, aux);
}
}
static void
-subcommand_statistics (int *keywords, pspp_linreg_cache * c)
+subcommand_statistics (int *keywords, linreg * c, void *aux)
{
/*
The order here must match the order in which the STATISTICS
keywords[r] = 1;
}
}
- statistics_keyword_output (reg_stats_r, keywords[r], c);
- statistics_keyword_output (reg_stats_anova, keywords[anova], c);
- statistics_keyword_output (reg_stats_coeff, keywords[coeff], c);
- statistics_keyword_output (reg_stats_outs, keywords[outs], c);
- statistics_keyword_output (reg_stats_zpp, keywords[zpp], c);
- statistics_keyword_output (reg_stats_label, keywords[label], c);
- statistics_keyword_output (reg_stats_sha, keywords[sha], c);
- statistics_keyword_output (reg_stats_ci, keywords[ci], c);
- statistics_keyword_output (reg_stats_f, keywords[f], c);
- statistics_keyword_output (reg_stats_bcov, keywords[bcov], c);
- statistics_keyword_output (reg_stats_ses, keywords[ses], c);
- statistics_keyword_output (reg_stats_xtx, keywords[xtx], c);
- statistics_keyword_output (reg_stats_collin, keywords[collin], c);
- statistics_keyword_output (reg_stats_tol, keywords[tol], c);
- statistics_keyword_output (reg_stats_selection, keywords[selection], c);
+ statistics_keyword_output (reg_stats_r, keywords[r], c, aux);
+ statistics_keyword_output (reg_stats_anova, keywords[anova], c, aux);
+ statistics_keyword_output (reg_stats_coeff, keywords[coeff], c, aux);
+ statistics_keyword_output (reg_stats_outs, keywords[outs], c, aux);
+ statistics_keyword_output (reg_stats_zpp, keywords[zpp], c, aux);
+ statistics_keyword_output (reg_stats_label, keywords[label], c, aux);
+ statistics_keyword_output (reg_stats_sha, keywords[sha], c, aux);
+ statistics_keyword_output (reg_stats_ci, keywords[ci], c, aux);
+ statistics_keyword_output (reg_stats_f, keywords[f], c, aux);
+ statistics_keyword_output (reg_stats_bcov, keywords[bcov], c, aux);
+ statistics_keyword_output (reg_stats_ses, keywords[ses], c, aux);
+ statistics_keyword_output (reg_stats_xtx, keywords[xtx], c, aux);
+ statistics_keyword_output (reg_stats_collin, keywords[collin], c, aux);
+ statistics_keyword_output (reg_stats_tol, keywords[tol], c, aux);
+ statistics_keyword_output (reg_stats_selection, keywords[selection], c, aux);
}
/*
if (t->trns_id == t->n_trns)
{
- result = pspp_linreg_cache_free (t->c);
+ result = linreg_free (t->c);
}
free (t);
Gets the predicted values.
*/
static int
-regression_trns_pred_proc (void *t_, struct ccase *c,
+regression_trns_pred_proc (void *t_, struct ccase **c,
casenumber case_idx UNUSED)
{
size_t i;
size_t n_vals;
struct reg_trns *trns = t_;
- pspp_linreg_cache *model;
+ linreg *model;
union value *output = NULL;
- const union value **vals = NULL;
+ const union value *tmp;
+ double *vals;
const struct variable **vars = NULL;
assert (trns != NULL);
assert (model->depvar != NULL);
assert (model->pred != NULL);
- vars = xnmalloc (model->n_coeffs, sizeof (*vars));
- n_vals = (*model->get_vars) (model, vars);
-
+ vars = linreg_get_vars (model);
+ n_vals = linreg_n_coeffs (model);
vals = xnmalloc (n_vals, sizeof (*vals));
- output = case_data_rw (c, model->pred);
- assert (output != NULL);
+ *c = case_unshare (*c);
+
+ output = case_data_rw (*c, model->pred);
for (i = 0; i < n_vals; i++)
{
- vals[i] = case_data (c, vars[i]);
+ tmp = case_data (*c, vars[i]);
+ vals[i] = tmp->f;
}
- output->f = (*model->predict) ((const struct variable **) vars,
- vals, model, n_vals);
+ output->f = linreg_predict (model, vals, n_vals);
free (vals);
- free (vars);
return TRNS_CONTINUE;
}
Gets the residuals.
*/
static int
-regression_trns_resid_proc (void *t_, struct ccase *c,
+regression_trns_resid_proc (void *t_, struct ccase **c,
casenumber case_idx UNUSED)
{
size_t i;
size_t n_vals;
struct reg_trns *trns = t_;
- pspp_linreg_cache *model;
+ linreg *model;
union value *output = NULL;
- const union value **vals = NULL;
- const union value *obs = NULL;
+ const union value *tmp;
+ double *vals = NULL;
+ double obs;
const struct variable **vars = NULL;
assert (trns != NULL);
assert (model->depvar != NULL);
assert (model->resid != NULL);
- vars = xnmalloc (model->n_coeffs, sizeof (*vars));
- n_vals = (*model->get_vars) (model, vars);
+ vars = linreg_get_vars (model);
+ n_vals = linreg_n_coeffs (model);
vals = xnmalloc (n_vals, sizeof (*vals));
- output = case_data_rw (c, model->resid);
+ *c = case_unshare (*c);
+ output = case_data_rw (*c, model->resid);
assert (output != NULL);
for (i = 0; i < n_vals; i++)
{
- vals[i] = case_data (c, vars[i]);
+ tmp = case_data (*c, vars[i]);
+ vals[i] = tmp->f;
}
- obs = case_data (c, model->depvar);
- output->f = (*model->residual) ((const struct variable **) vars,
- vals, obs, model, n_vals);
+ tmp = case_data (*c, model->depvar);
+ obs = tmp->f;
+ output->f = linreg_residual (model, obs, vals, n_vals);
free (vals);
- free (vars);
+
return TRNS_CONTINUE;
}
}
static void
-reg_get_name (const struct dictionary *dict, char name[LONG_NAME_LEN],
- const char prefix[LONG_NAME_LEN])
+reg_get_name (const struct dictionary *dict, char name[VAR_NAME_LEN],
+ const char prefix[VAR_NAME_LEN])
{
int i = 1;
- snprintf (name, LONG_NAME_LEN, "%s%d", prefix, i);
+ snprintf (name, VAR_NAME_LEN, "%s%d", prefix, i);
while (!try_name (dict, name))
{
i++;
- snprintf (name, LONG_NAME_LEN, "%s%d", prefix, i);
+ snprintf (name, VAR_NAME_LEN, "%s%d", prefix, i);
}
}
static void
reg_save_var (struct dataset *ds, const char *prefix, trns_proc_func * f,
- pspp_linreg_cache * c, struct variable **v, int n_trns)
+ linreg * c, struct variable **v, int n_trns)
{
struct dictionary *dict = dataset_dict (ds);
static int trns_index = 1;
- char name[LONG_NAME_LEN];
+ char name[VAR_NAME_LEN];
struct variable *new_var;
struct reg_trns *t = NULL;
add_transformation (ds, f, regression_trns_free, t);
trns_index++;
}
-
static void
-subcommand_save (struct dataset *ds, int save, pspp_linreg_cache ** models)
+subcommand_save (struct dataset *ds, int save, linreg ** models)
{
- pspp_linreg_cache **lc;
+ linreg **lc;
int n_trns = 0;
int i;
- assert (models != NULL);
-
if (save)
{
/* Count the number of transformations we will need. */
for (lc = models; lc < models + cmd.n_dependent; lc++)
{
- assert (*lc != NULL);
- assert ((*lc)->depvar != NULL);
- if (cmd.a_save[REGRESSION_SV_RESID])
- {
- reg_save_var (ds, "RES", regression_trns_resid_proc, *lc,
- &(*lc)->resid, n_trns);
- }
- if (cmd.a_save[REGRESSION_SV_PRED])
+ if (*lc != NULL)
{
- reg_save_var (ds, "PRED", regression_trns_pred_proc, *lc,
- &(*lc)->pred, n_trns);
+ if ((*lc)->depvar != NULL)
+ {
+ if (cmd.a_save[REGRESSION_SV_RESID])
+ {
+ reg_save_var (ds, "RES", regression_trns_resid_proc, *lc,
+ &(*lc)->resid, n_trns);
+ }
+ if (cmd.a_save[REGRESSION_SV_PRED])
+ {
+ reg_save_var (ds, "PRED", regression_trns_pred_proc, *lc,
+ &(*lc)->pred, n_trns);
+ }
+ }
}
}
}
{
if (*lc != NULL)
{
- pspp_linreg_cache_free (*lc);
- }
- }
- }
-}
-
-static int
-reg_inserted (const struct variable *v, struct variable **varlist, int n_vars)
-{
- int i;
-
- for (i = 0; i < n_vars; i++)
- {
- if (v == varlist[i])
- {
- return 1;
- }
- }
- return 0;
-}
-
-static void
-reg_print_categorical_encoding (FILE * fp, pspp_linreg_cache * c)
-{
- int i;
- int n_vars = 0;
- struct variable **varlist;
-
- fprintf (fp, "%s", reg_export_categorical_encode_1);
-
- varlist = xnmalloc (c->n_indeps, sizeof (*varlist));
- for (i = 1; i < c->n_indeps; i++) /* c->coeff[0] is the intercept. */
- {
- struct pspp_coeff *coeff = c->coeff[i];
- const struct variable *v = pspp_coeff_get_var (coeff, 0);
- if (var_is_alpha (v))
- {
- if (!reg_inserted (v, varlist, n_vars))
- {
- fprintf (fp, "struct pspp_reg_categorical_variable %s;\n\t",
- var_get_name (v));
- varlist[n_vars] = (struct variable *) v;
- n_vars++;
- }
- }
- }
- fprintf (fp, "int n_vars = %d;\n\t", n_vars);
- fprintf (fp, "struct pspp_reg_categorical_variable *varlist[%d] = {",
- n_vars);
- for (i = 0; i < n_vars - 1; i++)
- {
- fprintf (fp, "&%s,\n\t\t", var_get_name (varlist[i]));
- }
- fprintf (fp, "&%s};\n\t", var_get_name (varlist[i]));
-
- for (i = 0; i < n_vars; i++)
- {
- int n_categories = cat_get_n_categories (varlist[i]);
- int j;
-
- fprintf (fp, "%s.name = \"%s\";\n\t",
- var_get_name (varlist[i]), var_get_name (varlist[i]));
- fprintf (fp, "%s.n_vals = %d;\n\t",
- var_get_name (varlist[i]), n_categories);
-
- for (j = 0; j < n_categories; j++)
- {
- const union value *val = cat_subscript_to_value (j, varlist[i]);
- fprintf (fp, "%s.values[%d] = \"%s\";\n\t",
- var_get_name (varlist[i]), j,
- var_get_value_name (varlist[i], val));
- }
- }
- fprintf (fp, "%s", reg_export_categorical_encode_2);
-}
-
-static void
-reg_print_depvars (FILE * fp, pspp_linreg_cache * c)
-{
- int i;
- struct pspp_coeff *coeff;
- const struct variable *v;
-
- fprintf (fp, "char *model_depvars[%d] = {", c->n_indeps);
- for (i = 1; i < c->n_indeps; i++)
- {
- coeff = c->coeff[i];
- v = pspp_coeff_get_var (coeff, 0);
- fprintf (fp, "\"%s\",\n\t\t", var_get_name (v));
- }
- coeff = c->coeff[i];
- v = pspp_coeff_get_var (coeff, 0);
- fprintf (fp, "\"%s\"};\n\t", var_get_name (v));
-}
-static void
-reg_print_getvar (FILE * fp, pspp_linreg_cache * c)
-{
- fprintf (fp, "static int\npspp_reg_getvar (char *v_name)\n{\n\t");
- fprintf (fp, "int i;\n\tint n_vars = %d;\n\t", c->n_indeps);
- reg_print_depvars (fp, c);
- fprintf (fp, "for (i = 0; i < n_vars; i++)\n\t{\n\t\t");
- fprintf (fp,
- "if (strncmp (v_name, model_depvars[i], PSPP_REG_MAXLEN) == 0)\n\t\t{\n\t\t\t");
- fprintf (fp, "return i;\n\t\t}\n\t}\n}\n");
-}
-static int
-reg_has_categorical (pspp_linreg_cache * c)
-{
- int i;
- const struct variable *v;
-
- for (i = 1; i < c->n_coeffs; i++)
- {
- v = pspp_coeff_get_var (c->coeff[i], 0);
- if (var_is_alpha (v))
- return 1;
- }
- return 0;
-}
-
-static void
-subcommand_export (int export, pspp_linreg_cache * c)
-{
- FILE *fp;
- size_t i;
- size_t j;
- int n_quantiles = 100;
- double tmp;
- struct pspp_coeff *coeff;
-
- if (export)
- {
- assert (c != NULL);
- assert (model_file != NULL);
- fp = fopen (fh_get_file_name (model_file), "w");
- assert (fp != NULL);
- fprintf (fp, "%s", reg_preamble);
- reg_print_getvar (fp, c);
- if (reg_has_categorical (c))
- {
- reg_print_categorical_encoding (fp, c);
- }
- fprintf (fp, "%s", reg_export_t_quantiles_1);
- for (i = 0; i < n_quantiles - 1; i++)
- {
- tmp = 0.5 + 0.005 * (double) i;
- fprintf (fp, "%.15e,\n\t\t",
- gsl_cdf_tdist_Pinv (tmp, c->n_obs - c->n_indeps));
- }
- fprintf (fp, "%.15e};\n\t",
- gsl_cdf_tdist_Pinv (.9995, c->n_obs - c->n_indeps));
- fprintf (fp, "%s", reg_export_t_quantiles_2);
- fprintf (fp, "%s", reg_mean_cmt);
- fprintf (fp, "double\npspp_reg_estimate (const double *var_vals,");
- fprintf (fp, "const char *var_names[])\n{\n\t");
- fprintf (fp, "double model_coeffs[%d] = {", c->n_indeps);
- for (i = 1; i < c->n_indeps; i++)
- {
- coeff = c->coeff[i];
- fprintf (fp, "%.15e,\n\t\t", coeff->estimate);
- }
- coeff = c->coeff[i];
- fprintf (fp, "%.15e};\n\t", coeff->estimate);
- coeff = c->coeff[0];
- fprintf (fp, "double estimate = %.15e;\n\t", coeff->estimate);
- fprintf (fp, "int i;\n\tint j;\n\n\t");
- fprintf (fp, "for (i = 0; i < %d; i++)\n\t", c->n_indeps);
- fprintf (fp, "%s", reg_getvar);
- fprintf (fp, "const double cov[%d][%d] = {\n\t", c->n_coeffs,
- c->n_coeffs);
- for (i = 0; i < c->cov->size1 - 1; i++)
- {
- fprintf (fp, "{");
- for (j = 0; j < c->cov->size2 - 1; j++)
- {
- fprintf (fp, "%.15e, ", gsl_matrix_get (c->cov, i, j));
+ linreg_free (*lc);
}
- fprintf (fp, "%.15e},\n\t", gsl_matrix_get (c->cov, i, j));
}
- fprintf (fp, "{");
- for (j = 0; j < c->cov->size2 - 1; j++)
- {
- fprintf (fp, "%.15e, ",
- gsl_matrix_get (c->cov, c->cov->size1 - 1, j));
- }
- fprintf (fp, "%.15e}\n\t",
- gsl_matrix_get (c->cov, c->cov->size1 - 1, c->cov->size2 - 1));
- fprintf (fp, "};\n\tint n_vars = %d;\n\tint i;\n\tint j;\n\t",
- c->n_indeps);
- fprintf (fp, "double unshuffled_vals[%d];\n\t", c->n_indeps);
- fprintf (fp, "%s", reg_variance);
- fprintf (fp, "%s", reg_export_confidence_interval);
- tmp = c->mse * c->mse;
- fprintf (fp, "%s %.15e", reg_export_prediction_interval_1, tmp);
- fprintf (fp, "%s %.15e", reg_export_prediction_interval_2, tmp);
- fprintf (fp, "%s", reg_export_prediction_interval_3);
- fclose (fp);
- fp = fopen ("pspp_model_reg.h", "w");
- fprintf (fp, "%s", reg_header);
- fclose (fp);
}
}
-static int
-regression_custom_export (struct lexer *lexer, struct dataset *ds UNUSED,
- struct cmd_regression *cmd UNUSED, void *aux UNUSED)
-{
- /* 0 on failure, 1 on success, 2 on failure that should result in syntax error */
- if (!lex_force_match (lexer, '('))
- return 0;
-
- if (lex_match (lexer, '*'))
- model_file = NULL;
- else
- {
- model_file = fh_parse (lexer, FH_REF_FILE);
- if (model_file == NULL)
- return 0;
- }
-
- if (!lex_force_match (lexer, ')'))
- return 0;
-
- return 1;
-}
-
int
cmd_regression (struct lexer *lexer, struct dataset *ds)
{
struct casegrouper *grouper;
struct casereader *group;
+ linreg **models;
bool ok;
size_t i;
if (!parse_regression (lexer, ds, &cmd, NULL))
- return CMD_FAILURE;
+ {
+ return CMD_FAILURE;
+ }
models = xnmalloc (cmd.n_dependent, sizeof *models);
for (i = 0; i < cmd.n_dependent; i++)
/* Data pass. */
grouper = casegrouper_create_splits (proc_open (ds), dataset_dict (ds));
while (casegrouper_get_next_group (grouper, &group))
- run_regression (group, &cmd, ds);
+ run_regression (group, &cmd, ds, models);
ok = casegrouper_destroy (grouper);
ok = proc_commit (ds) && ok;
subcommand_save (ds, cmd.sbc_save, models);
free (v_variables);
free (models);
+ free_regression (&cmd);
+
return ok ? CMD_SUCCESS : CMD_FAILURE;
}
{
const struct dictionary *dict = dataset_dict (ds);
- lex_match (lexer, '=');
+ lex_match (lexer, T_EQUALS);
if ((lex_token (lexer) != T_ID
- || dict_lookup_var (dict, lex_tokid (lexer)) == NULL)
+ || dict_lookup_var (dict, lex_tokcstr (lexer)) == NULL)
&& lex_token (lexer) != T_ALL)
return 2;
for (i = 0; i < n_variables; i++)
if (!is_depvar (i, depvar))
indep_vars[n_indep_vars++] = v_variables[i];
- if ((n_indep_vars < 2) && is_depvar (0, depvar))
+ if ((n_indep_vars < 1) && is_depvar (0, depvar))
{
/*
There is only one independent variable, and it is the same
as the dependent variable. Print a warning and continue.
*/
msg (SE,
- gettext ("The dependent variable is equal to the independent variable. The least sequares line is therefore Y=X. Standard errors and related statistics may be meaningless."));
+ gettext ("The dependent variable is equal to the independent variable."
+ "The least squares line is therefore Y=X."
+ "Standard errors and related statistics may be meaningless."));
n_indep_vars = 1;
indep_vars[0] = v_variables[0];
}
return n_indep_vars;
}
-
-/* Encode categorical variables.
- Returns number of valid cases. */
-static int
-prepare_categories (struct casereader *input,
- const struct variable **vars, size_t n_vars,
- struct moments_var *mom)
+static double
+fill_covariance (gsl_matrix *cov, struct covariance *all_cov,
+ const struct variable **vars,
+ size_t n_vars, const struct variable *dep_var,
+ const struct variable **all_vars, size_t n_all_vars,
+ double *means)
{
- int n_data;
- struct ccase c;
size_t i;
-
- assert (vars != NULL);
- assert (mom != NULL);
-
- for (i = 0; i < n_vars; i++)
- if (var_is_alpha (vars[i]))
- cat_stored_values_create (vars[i]);
-
- n_data = 0;
- for (; casereader_read (input, &c); case_destroy (&c))
+ size_t j;
+ size_t dep_subscript;
+ size_t *rows;
+ const gsl_matrix *ssizes;
+ const gsl_matrix *cm;
+ const gsl_matrix *mean_matrix;
+ const gsl_matrix *ssize_matrix;
+ double result = 0.0;
+
+ cm = covariance_calculate_unnormalized (all_cov);
+ rows = xnmalloc (cov->size1 - 1, sizeof (*rows));
+
+ for (i = 0; i < n_all_vars; i++)
{
- /*
- The second condition ensures the program will run even if
- there is only one variable to act as both explanatory and
- response.
- */
- for (i = 0; i < n_vars; i++)
+ for (j = 0; j < n_vars; j++)
{
- const union value *val = case_data (&c, vars[i]);
- if (var_is_alpha (vars[i]))
- cat_value_update (vars[i], val);
- else
- moments1_add (mom[i].m, val->f, 1.0);
+ if (vars[j] == all_vars[i])
+ {
+ rows[j] = i;
+ }
+ }
+ if (all_vars[i] == dep_var)
+ {
+ dep_subscript = i;
}
- n_data++;
}
- casereader_destroy (input);
-
- return n_data;
+ mean_matrix = covariance_moments (all_cov, MOMENT_MEAN);
+ ssize_matrix = covariance_moments (all_cov, MOMENT_NONE);
+ for (i = 0; i < cov->size1 - 1; i++)
+ {
+ means[i] = gsl_matrix_get (mean_matrix, rows[i], 0)
+ / gsl_matrix_get (ssize_matrix, rows[i], 0);
+ for (j = 0; j < cov->size2 - 1; j++)
+ {
+ gsl_matrix_set (cov, i, j, gsl_matrix_get (cm, rows[i], rows[j]));
+ gsl_matrix_set (cov, j, i, gsl_matrix_get (cm, rows[j], rows[i]));
+ }
+ }
+ means[cov->size1 - 1] = gsl_matrix_get (mean_matrix, dep_subscript, 0)
+ / gsl_matrix_get (ssize_matrix, dep_subscript, 0);
+ ssizes = covariance_moments (all_cov, MOMENT_NONE);
+ result = gsl_matrix_get (ssizes, dep_subscript, rows[0]);
+ for (i = 0; i < cov->size1 - 1; i++)
+ {
+ gsl_matrix_set (cov, i, cov->size1 - 1,
+ gsl_matrix_get (cm, rows[i], dep_subscript));
+ gsl_matrix_set (cov, cov->size1 - 1, i,
+ gsl_matrix_get (cm, rows[i], dep_subscript));
+ if (result > gsl_matrix_get (ssizes, rows[i], dep_subscript))
+ {
+ result = gsl_matrix_get (ssizes, rows[i], dep_subscript);
+ }
+ }
+ gsl_matrix_set (cov, cov->size1 - 1, cov->size1 - 1,
+ gsl_matrix_get (cm, dep_subscript, dep_subscript));
+ free (rows);
+ gsl_matrix_free (cm);
+ return result;
}
-
-static void
-coeff_init (pspp_linreg_cache * c, struct design_matrix *dm)
+static size_t
+get_n_all_vars (struct cmd_regression *cmd)
{
- c->coeff = xnmalloc (dm->m->size2 + 1, sizeof (*c->coeff));
- c->coeff[0] = xmalloc (sizeof (*(c->coeff[0]))); /* The first coefficient is the intercept. */
- c->coeff[0]->v_info = NULL; /* Intercept has no associated variable. */
- pspp_coeff_init (c->coeff + 1, dm);
-}
+ size_t result = n_variables;
+ size_t i;
+ size_t j;
-/*
- Put the moments in the linreg cache.
- */
+ result += cmd->n_dependent;
+ for (i = 0; i < cmd->n_dependent; i++)
+ {
+ for (j = 0; j < n_variables; j++)
+ {
+ if (v_variables[j] == cmd->v_dependent[i])
+ {
+ result--;
+ }
+ }
+ }
+ return result;
+}
static void
-compute_moments (pspp_linreg_cache * c, struct moments_var *mom,
- struct design_matrix *dm, size_t n)
+fill_all_vars (const struct variable **vars, struct cmd_regression *cmd)
{
size_t i;
size_t j;
- double weight;
- double mean;
- double variance;
- double skewness;
- double kurtosis;
- /*
- Scan the variable names in the columns of the design matrix.
- When we find the variable we need, insert its mean in the cache.
- */
- for (i = 0; i < dm->m->size2; i++)
+ bool absent;
+
+ for (i = 0; i < n_variables; i++)
{
- for (j = 0; j < n; j++)
+ vars[i] = v_variables[i];
+ }
+ for (i = 0; i < cmd->n_dependent; i++)
+ {
+ absent = true;
+ for (j = 0; j < n_variables; j++)
{
- if (design_matrix_col_to_var (dm, i) == (mom + j)->v)
+ if (cmd->v_dependent[i] == v_variables[j])
{
- moments1_calculate ((mom + j)->m, &weight, &mean, &variance,
- &skewness, &kurtosis);
- gsl_vector_set (c->indep_means, i, mean);
- gsl_vector_set (c->indep_std, i, sqrt (variance));
+ absent = false;
+ break;
}
}
+ if (absent)
+ {
+ vars[i + n_variables] = cmd->v_dependent[i];
+ }
}
}
-
static bool
run_regression (struct casereader *input, struct cmd_regression *cmd,
- struct dataset *ds)
+ struct dataset *ds, linreg **models)
{
size_t i;
int n_indep = 0;
int k;
- struct ccase c;
- const struct variable **indep_vars;
- struct design_matrix *X;
- struct moments_var *mom;
- gsl_vector *Y;
-
- pspp_linreg_opts lopts;
+ double n_data;
+ double *means;
+ struct ccase *c;
+ struct covariance *cov;
+ const struct variable **vars;
+ const struct variable **all_vars;
+ const struct variable *dep_var;
+ struct casereader *reader;
+ const struct dictionary *dict;
+ size_t n_all_vars;
assert (models != NULL);
- if (!casereader_peek (input, 0, &c))
- return true;
- output_split_file_values (ds, &c);
- case_destroy (&c);
-
- if (!v_variables)
- {
- dict_get_vars (dataset_dict (ds), &v_variables, &n_variables,
- 1u << DC_SYSTEM);
- }
-
- for (i = 0; i < cmd->n_dependent; i++)
+ for (i = 0; i < n_variables; i++)
{
- if (!var_is_numeric (cmd->v_dependent[i]))
+ if (!var_is_numeric (v_variables[i]))
{
- msg (SE, _("Dependent variable must be numeric."));
+ msg (SE, _("REGRESSION requires numeric variables."));
return false;
}
}
- mom = xnmalloc (n_variables, sizeof (*mom));
- for (i = 0; i < n_variables; i++)
+ c = casereader_peek (input, 0);
+ if (c == NULL)
{
- (mom + i)->m = moments1_create (MOMENT_VARIANCE);
- (mom + i)->v = v_variables[i];
+ casereader_destroy (input);
+ return true;
}
- lopts.get_depvar_mean_std = 1;
-
- lopts.get_indep_mean_std = xnmalloc (n_variables, sizeof (int));
- indep_vars = xnmalloc (n_variables, sizeof *indep_vars);
+ output_split_file_values (ds, c);
+ case_unref (c);
+ dict = dataset_dict (ds);
+ if (!v_variables)
+ {
+ dict_get_vars (dict, &v_variables, &n_variables, 0);
+ }
+ n_all_vars = get_n_all_vars (cmd);
+ all_vars = xnmalloc (n_all_vars, sizeof (*all_vars));
+ fill_all_vars (all_vars, cmd);
+ vars = xnmalloc (n_variables, sizeof (*vars));
+ means = xnmalloc (n_all_vars, sizeof (*means));
+ cov = covariance_1pass_create (n_all_vars, all_vars,
+ dict_get_weight (dict), MV_ANY);
+
+ reader = casereader_clone (input);
+ reader = casereader_create_filter_missing (reader, v_variables, n_variables,
+ MV_ANY, NULL, NULL);
+ for (; (c = casereader_read (reader)) != NULL; case_unref (c))
+ {
+ covariance_accumulate (cov, c);
+ }
+
for (k = 0; k < cmd->n_dependent; k++)
{
- const struct variable *dep_var;
- struct casereader *reader;
- casenumber row;
- struct ccase c;
- size_t n_data; /* Number of valid cases. */
-
+ gsl_matrix *this_cm;
dep_var = cmd->v_dependent[k];
- n_indep = identify_indep_vars (indep_vars, dep_var);
- reader = casereader_clone (input);
- reader = casereader_create_filter_missing (reader, indep_vars, n_indep,
- MV_ANY, NULL);
- reader = casereader_create_filter_missing (reader, &dep_var, 1,
- MV_ANY, NULL);
- n_data = prepare_categories (casereader_clone (reader),
- indep_vars, n_indep, mom);
-
- if ((n_data > 0) && (n_indep > 0))
+ n_indep = identify_indep_vars (vars, dep_var);
+
+ this_cm = gsl_matrix_alloc (n_indep + 1, n_indep + 1);
+ n_data = fill_covariance (this_cm, cov, vars, n_indep,
+ dep_var, all_vars, n_all_vars, means);
+ models[k] = linreg_alloc (dep_var, (const struct variable **) vars,
+ n_data, n_indep);
+ models[k]->depvar = dep_var;
+ for (i = 0; i < n_indep; i++)
+ {
+ linreg_set_indep_variable_mean (models[k], i, means[i]);
+ }
+ linreg_set_depvar_mean (models[k], means[i]);
+ /*
+ For large data sets, use QR decomposition.
+ */
+ if (n_data > sqrt (n_indep) && n_data > REG_LARGE_DATA)
+ {
+ models[k]->method = LINREG_QR;
+ }
+
+ if (n_data > 0)
{
- Y = gsl_vector_alloc (n_data);
- X =
- design_matrix_create (n_indep,
- (const struct variable **) indep_vars,
- n_data);
- for (i = 0; i < X->m->size2; i++)
- {
- lopts.get_indep_mean_std[i] = 1;
- }
- models[k] = pspp_linreg_cache_alloc (X->m->size1, X->m->size2);
- models[k]->indep_means = gsl_vector_alloc (X->m->size2);
- models[k]->indep_std = gsl_vector_alloc (X->m->size2);
- models[k]->depvar = dep_var;
- /*
- For large data sets, use QR decomposition.
- */
- if (n_data > sqrt (n_indep) && n_data > REG_LARGE_DATA)
- {
- models[k]->method = PSPP_LINREG_QR;
- }
-
- /*
- The second pass fills the design matrix.
- */
- reader = casereader_create_counter (reader, &row, -1);
- for (; casereader_read (reader, &c); case_destroy (&c))
- {
- for (i = 0; i < n_indep; ++i)
- {
- const struct variable *v = indep_vars[i];
- const union value *val = case_data (&c, v);
- if (var_is_alpha (v))
- design_matrix_set_categorical (X, row, v, val);
- else
- design_matrix_set_numeric (X, row, v, val);
- }
- gsl_vector_set (Y, row, case_num (&c, dep_var));
- }
- /*
- Now that we know the number of coefficients, allocate space
- and store pointers to the variables that correspond to the
- coefficients.
- */
- coeff_init (models[k], X);
-
/*
- Find the least-squares estimates and other statistics.
- */
- pspp_linreg ((const gsl_vector *) Y, X->m, &lopts, models[k]);
- compute_moments (models[k], mom, X, n_variables);
-
+ Find the least-squares estimates and other statistics.
+ */
+ linreg_fit (this_cm, models[k]);
+
if (!taint_has_tainted_successor (casereader_get_taint (input)))
{
- subcommand_statistics (cmd->a_statistics, models[k]);
- subcommand_export (cmd->sbc_export, models[k]);
+ subcommand_statistics (cmd->a_statistics, models[k], this_cm);
}
-
- gsl_vector_free (Y);
- design_matrix_destroy (X);
}
else
{
msg (SE,
gettext ("No valid data found. This command was skipped."));
+ linreg_free (models[k]);
+ models[k] = NULL;
}
- casereader_destroy (reader);
+ gsl_matrix_free (this_cm);
}
- free (indep_vars);
- free (lopts.get_indep_mean_std);
+
+ casereader_destroy (reader);
+ free (vars);
+ free (all_vars);
+ free (means);
casereader_destroy (input);
-
+ covariance_destroy (cov);
+
return true;
}