From: Ben Pfaff <blp@cs.stanford.edu>
Date: Wed, 10 Mar 2010 04:52:48 +0000 (-0800)
Subject: FREQUENCIES: Choose number of bins for histogram based on valid cases.
X-Git-Tag: v0.7.5~118
X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=c919bc2cf709771292738b0a64a07bd04b3417ae;p=pspp-builds.git

FREQUENCIES: Choose number of bins for histogram based on valid cases.

Until now, histograms have always had exactly 11 bins.  There is no
"correct" number of bins, but a fixed number of bins also seems less than
ideal.  Use Sturges' formula, instead, to choose the number of bins.

Reported by Erik Frebold <efrebold@interchange.ubc.ca>.
---

diff --git a/src/language/stats/frequencies.q b/src/language/stats/frequencies.q
index d00c90c9..523345be 100644
--- a/src/language/stats/frequencies.q
+++ b/src/language/stats/frequencies.q
@@ -1374,7 +1374,7 @@ freq_tab_to_hist (const struct freq_tab *ft, const struct variable *var)
   double x_max = -DBL_MAX;
 
   struct histogram *hist;
-  const double bins = 11;
+  int bins;
 
   struct hsh_iterator hi;
   struct hsh_table *fh = ft->data;
@@ -1390,6 +1390,11 @@ freq_tab_to_hist (const struct freq_tab *ft, const struct variable *var)
       if ( frq->value.f > x_max ) x_max = frq->value.f ;
     }
 
+  /* Sturges' formula. */
+  bins = ceil (log (ft->valid_cases) / log (2) + 1);
+  if (bins < 5)
+    bins = 5;
+
   hist = histogram_create (bins, x_min, x_max);
 
   for( i = 0 ; i < ft->n_valid ; ++i )