src/math/box-whisker.c

   1 /* PSPP - a program for statistical analysis.
   2    Copyright (C) 2008, 2009, 2011 Free Software Foundation, Inc.
   3
   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.
   8
   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.
  13
  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/>. */
  16
  17 #include <config.h>
  18
  19 #include "box-whisker.h"
  20
  21 #include <math.h>
  22 #include <float.h>
  23
  24 #include "data/case.h"
  25 #include "data/data-out.h"
  26 #include "data/val-type.h"
  27 #include "data/variable.h"
  28 #include "libpspp/assertion.h"
  29 #include "libpspp/cast.h"
  30 #include "libpspp/str.h"
  31 #include "math/order-stats.h"
  32 #include "math/tukey-hinges.h"
  33
  34 #include "gl/xalloc.h"
  35
  36 static void
  37 destroy (struct statistic *s)
  38 {
  39   struct box_whisker *bw = UP_CAST (s, struct box_whisker, parent.parent);
  40   struct order_stats *os = &bw->parent;
  41   struct ll *ll;
  42
  43   for (ll = ll_head (&bw->outliers); ll != ll_null (&bw->outliers);)
  44     {
  45       struct outlier *e = ll_data (ll, struct outlier, ll);
  46
  47       ll = ll_next (ll);
  48
  49       ds_destroy (&e->label);
  50       free (e);
  51     }
  52
  53   free (os->k);
  54   free (s);
  55 };
  56
  57
  58 static void
  59 acc (struct statistic *s, const struct ccase *cx,
  60      double c UNUSED, double cc UNUSED, double y)
  61 {
  62   struct box_whisker *bw = UP_CAST (s, struct box_whisker, parent.parent);
  63   bool extreme;
  64   struct outlier *o;
  65
  66   if (y > bw->hinges[2] + bw->step) /* Upper outlier */
  67     {
  68       extreme = (y > bw->hinges[2] + 2 * bw->step) ;
  69     }
  70
  71   else if (y < bw->hinges[0] - bw->step) /* Lower outlier */
  72     {
  73       extreme = (y < bw->hinges[0] - 2 * bw->step) ;
  74     }
  75
  76   else /* Not an outlier */
  77     {
  78       if (bw->whiskers[0] == SYSMIS)
  79         bw->whiskers[0] = y;
  80
  81       if (y > bw->whiskers[1])
  82         bw->whiskers[1] = y;
  83
  84       return;
  85     }
  86
  87   /* y is an outlier */
  88
  89   o = xzalloc (sizeof *o) ;
  90   o->value = y;
  91   o->extreme = extreme;
  92   ds_init_empty (&o->label);
  93
  94   if (bw->id_var)
  95     {
  96       char *s = data_out (case_data_idx (cx, bw->id_idx),
  97                            var_get_encoding (bw->id_var),
  98                            var_get_print_format (bw->id_var));
  99
 100       ds_put_cstr (&o->label, s);
 101       free (s);
 102     }
 103   else
 104     {
 105       ds_put_format (&o->label,
 106                      "%ld",
 107                      (casenumber) case_data_idx (cx, bw->id_idx)->f);
 108     }
 109
 110   ll_push_head (&bw->outliers, &o->ll);
 111 }
 112
 113 void
 114 box_whisker_whiskers (const struct box_whisker *bw, double whiskers[2])
 115 {
 116   whiskers[0] = bw->whiskers[0];
 117   whiskers[1] = bw->whiskers[1];
 118 }
 119
 120 void
 121 box_whisker_hinges (const struct box_whisker *bw, double hinges[3])
 122 {
 123   hinges[0] = bw->hinges[0];
 124   hinges[1] = bw->hinges[1];
 125   hinges[2] = bw->hinges[2];
 126 }
 127
 128 const struct ll_list *
 129 box_whisker_outliers (const struct box_whisker *bw)
 130 {
 131   return &bw->outliers;
 132 }
 133
 134 /*
 135   Create a box_whisker struct, suitable for generating a boxplot.
 136
 137   TH are the tukey hinges of the dataset.
 138
 139   id_idx is the index into the casereader which will be used to label
 140   outliers.
 141   id_var is the variable from which that label came, or NULL
 142 */
 143 struct box_whisker *
 144 box_whisker_create (const struct tukey_hinges *th,
 145                     size_t id_idx, const struct variable *id_var)
 146 {
 147   struct box_whisker *w = xzalloc (sizeof (*w));
 148   struct order_stats *os = &w->parent;
 149   struct statistic *stat = &os->parent;
 150
 151   os->n_k = 0;
 152
 153   stat->destroy = destroy;
 154   stat->accumulate = acc;
 155
 156   tukey_hinges_calculate (th, w->hinges);
 157
 158   w->id_idx = id_idx;
 159   w->id_var = id_var;
 160
 161   w->step = (w->hinges[2] - w->hinges[0]) * 1.5;
 162
 163   w->whiskers[1] = w->hinges[2];
 164   w->whiskers[0] = SYSMIS;
 165
 166   ll_init (&w->outliers);
 167
 168   return w;
 169 }