-/* PSPP - computes sample statistics.
- Copyright (C) 2006, 2007 Free Software Foundation, Inc.
+/* PSPP - a program for statistical analysis.
+ Copyright (C) 2006, 2007, 2009 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 the Free Software Foundation; either version 2 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 3 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
- 02110-1301, USA. */
+ along with this program. If not, see <http://www.gnu.org/licenses/>. */
#include <config.h>
#include <stdlib.h>
#include <math.h>
+#include <data/format.h>
#include <data/case.h>
#include <data/casereader.h>
#include <data/dictionary.h>
#include <data/variable.h>
#include <language/stats/freq.h>
#include <language/stats/npar.h>
-#include <libpspp/alloc.h>
#include <libpspp/assertion.h>
#include <libpspp/compiler.h>
#include <libpspp/hash.h>
#include <gsl/gsl_cdf.h>
+#include "xalloc.h"
+
#include "gettext.h"
#define _(msgid) gettext (msgid)
{
bool warn = true;
float i_d;
- struct ccase c;
+ struct ccase *c;
struct hsh_table *freq_hash =
hsh_create (4, compare_freq, hash_freq,
hsh_insert (freq_hash, fr);
}
- while (casereader_read (input, &c))
+ while ((c = casereader_read (input)) != NULL)
{
union value obs_value;
struct freq **existing_fr;
struct freq *fr = xmalloc(sizeof (*fr));
- fr->value = case_data (&c, var);
+ fr->value = case_data (c, var);
- fr->count = dict_get_case_weight (dict, &c, &warn);
+ fr->count = dict_get_case_weight (dict, c, &warn);
obs_value.f = trunc (fr->value->f);
if ( obs_value.f < lo || obs_value.f > hi)
{
free (fr);
- case_destroy (&c);
+ case_unref (c);
continue;
}
(*existing_fr)->count += fr->count;
free (fr);
- case_destroy (&c);
+ case_unref (c);
}
if (casereader_destroy (input))
return freq_hash;
const struct variable *var)
{
bool warn = true;
- struct ccase c;
+ struct ccase *c;
struct hsh_table *freq_hash =
hsh_create (4, compare_freq, hash_freq,
free_freq_mutable_hash,
(void *) var);
- for (; casereader_read (input, &c); case_destroy (&c))
+ for (; (c = casereader_read (input)) != NULL; case_unref (c))
{
struct freq **existing_fr;
struct freq *fr = xmalloc(sizeof (*fr));
- fr->value = case_data (&c, var);
+ fr->value = case_data (c, var);
- fr->count = dict_get_case_weight (dict, &c, &warn);
+ fr->count = dict_get_case_weight (dict, c, &warn);
existing_fr = (struct freq **) hsh_probe (freq_hash, fr);
if ( *existing_fr)
test->n_expected, n_cells,
var_get_name (var)
);
+ hsh_destroy (*freq_hash);
+ *freq_hash = NULL;
return NULL;
}
}
for ( i = test->lo ; i <= test->hi ; ++i )
- tab_float (table, 0, 2 + i - test->lo,
- TAB_LEFT, 1 + i - test->lo, 8, 0);
+ tab_fixed (table, 0, 2 + i - test->lo,
+ TAB_LEFT, 1 + i - test->lo, 8, 0);
tab_headers (table, 1, 0, 2, 0);
chisquare_execute (const struct dataset *ds,
struct casereader *input,
enum mv_class exclude,
- const struct npar_test *test)
+ const struct npar_test *test,
+ bool exact UNUSED,
+ double timer UNUSED)
{
const struct dictionary *dict = dataset_dict (ds);
int v, i;
struct chisquare_test *cst = (struct chisquare_test *) test;
int n_cells = 0;
double total_expected = 0.0;
+ const struct variable *wvar = dict_get_weight (dict);
+ const struct fmt_spec *wfmt = wvar ?
+ var_get_print_format (wvar) : & F_8_0;
double *df = xzalloc (sizeof (*df) * ost->n_vars);
double *xsq = xzalloc (sizeof (*df) * ost->n_vars);
struct hsh_table *freq_hash = NULL;
struct casereader *reader =
casereader_create_filter_missing (casereader_clone (input),
- &ost->vars[v], 1, exclude, NULL);
+ &ost->vars[v], 1, exclude,
+ NULL, NULL);
struct tab_table *freq_table =
create_variable_frequency_table(dict, reader, cst, v, &freq_hash);
xsq[v] = 0.0;
for ( i = 0 ; i < n_cells ; ++i )
{
+ struct string str;
double exp;
const union value *observed_value = ff[i]->value;
+ ds_init_empty (&str);
+ var_append_value_name (ost->vars[v], observed_value, &str);
+
/* The key */
- tab_text (freq_table, 0, i + 1, TAB_LEFT,
- var_get_value_name (ost->vars[v], observed_value));
+ tab_text (freq_table, 0, i + 1, TAB_LEFT, ds_cstr (&str));
+ ds_destroy (&str);
+
/* The observed N */
- tab_float (freq_table, 1, i + 1, TAB_NONE,
- ff[i]->count, 8, 0);
+ tab_double (freq_table, 1, i + 1, TAB_NONE,
+ ff[i]->count, wfmt);
if ( cst->n_expected > 0 )
exp = cst->expected[i] * total_obs / total_expected ;
else
exp = total_obs / (double) n_cells;
- tab_float (freq_table, 2, i + 1, TAB_NONE,
- exp, 8, 2);
+ tab_double (freq_table, 2, i + 1, TAB_NONE,
+ exp, NULL);
/* The residual */
- tab_float (freq_table, 3, i + 1, TAB_NONE,
- ff[i]->count - exp, 8, 2);
+ tab_double (freq_table, 3, i + 1, TAB_NONE,
+ ff[i]->count - exp, NULL);
xsq[v] += (ff[i]->count - exp) * (ff[i]->count - exp) / exp;
}
df[v] = n_cells - 1.0;
- tab_float (freq_table, 1, i + 1, TAB_NONE,
- total_obs, 8, 0);
+ tab_double (freq_table, 1, i + 1, TAB_NONE,
+ total_obs, wfmt);
tab_submit (freq_table);
double total_obs = 0.0;
struct casereader *reader =
casereader_create_filter_missing (casereader_clone (input),
- &ost->vars[v], 1, exclude, NULL);
+ &ost->vars[v], 1, exclude,
+ NULL, NULL);
struct hsh_table *freq_hash =
create_freq_hash_with_range (dict, reader,
ost->vars[v], cst->lo, cst->hi);
xsq[v] = 0.0;
for ( i = 0 ; i < hsh_count (freq_hash) ; ++i )
{
+ struct string str;
double exp;
const union value *observed_value = ff[i]->value;
+ ds_init_empty (&str);
+ var_append_value_name (ost->vars[v], observed_value, &str);
/* The key */
tab_text (freq_table, v * 4 + 1, i + 2 , TAB_LEFT,
- var_get_value_name (ost->vars[v], observed_value));
+ ds_cstr (&str));
+ ds_destroy (&str);
/* The observed N */
- tab_float (freq_table, v * 4 + 2, i + 2 , TAB_NONE,
- ff[i]->count, 8, 0);
+ tab_double (freq_table, v * 4 + 2, i + 2 , TAB_NONE,
+ ff[i]->count, wfmt);
if ( cst->n_expected > 0 )
exp = cst->expected[i] * total_obs / total_expected ;
exp = total_obs / (double) hsh_count (freq_hash);
/* The expected N */
- tab_float (freq_table, v * 4 + 3, i + 2 , TAB_NONE,
- exp, 8, 2);
+ tab_double (freq_table, v * 4 + 3, i + 2 , TAB_NONE,
+ exp, NULL);
/* The residual */
- tab_float (freq_table, v * 4 + 4, i + 2 , TAB_NONE,
- ff[i]->count - exp, 8, 2);
+ tab_double (freq_table, v * 4 + 4, i + 2 , TAB_NONE,
+ ff[i]->count - exp, NULL);
xsq[v] += (ff[i]->count - exp) * (ff[i]->count - exp) / exp;
}
- tab_float (freq_table, v * 4 + 2, tab_nr (freq_table) - 1, TAB_NONE,
- total_obs, 8, 0);
+ tab_double (freq_table, v * 4 + 2, tab_nr (freq_table) - 1, TAB_NONE,
+ total_obs, wfmt);
df[v] = n_cells - 1.0;
tab_text (stats_table, 1 + v, 0, TAB_CENTER, var_get_name (var));
- tab_float (stats_table, 1 + v, 1, TAB_NONE, xsq[v], 8,3);
- tab_float (stats_table, 1 + v, 2, TAB_NONE, df[v], 8,0);
+ tab_double (stats_table, 1 + v, 1, TAB_NONE, xsq[v], NULL);
+ tab_fixed (stats_table, 1 + v, 2, TAB_NONE, df[v], 8, 0);
- tab_float (stats_table, 1 + v, 3, TAB_NONE,
- gsl_cdf_chisq_Q (xsq[v], df[v]), 8,3);
+ tab_double (stats_table, 1 + v, 3, TAB_NONE,
+ gsl_cdf_chisq_Q (xsq[v], df[v]), NULL);
}
tab_submit (stats_table);
}