statsample 0.6.7 → 0.7.0

data/po/statsample.pot

@@ -6,8 +6,8 @@
 #, fuzzy
 msgid ""
 msgstr ""
-"Project-Id-Version: statsample 0.6.2\n"
-"POT-Creation-Date: 2010-02-11 18:44-0300\n"
+"Project-Id-Version: statsample 0.6.8\n"
+"POT-Creation-Date: 2010-03-25 17:55-0300\n"
 "PO-Revision-Date: 2009-08-04 15:36-0400\n"
 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
 "Language-Team: LANGUAGE <LL@li.org>\n"
@@ -16,135 +16,155 @@ msgstr ""
 "Content-Transfer-Encoding: 8bit\n"
 "Plural-Forms: nplurals=INTEGER; plural=EXPRESSION;\n"
 
-#: lib/statsample/regression/multiple/baseengine.rb:14
+#: lib/statsample/regression/multiple/baseengine.rb:23
 msgid "Multiple Regression:  %s over %s"
 msgstr ""
 
-#: lib/statsample/regression/multiple/baseengine.rb:150
+#: lib/statsample/regression/multiple/baseengine.rb:164
+#: lib/statsample/regression/multiple/matrixengine.rb:198
 msgid "Multiple Regression: "
 msgstr ""
 
-#: lib/statsample/regression/multiple/baseengine.rb:153
+#: lib/statsample/regression/multiple/baseengine.rb:167
+#: lib/statsample/regression/multiple/matrixengine.rb:201
 msgid "Engine: %s"
 msgstr ""
 
-#: lib/statsample/regression/multiple/baseengine.rb:154
+#: lib/statsample/regression/multiple/baseengine.rb:168
 msgid "Cases(listwise)=%d(%d)"
 msgstr ""
 
-#: lib/statsample/regression/multiple/baseengine.rb:158
+#: lib/statsample/regression/multiple/baseengine.rb:172
+#: lib/statsample/regression/multiple/matrixengine.rb:206
 msgid "Equation"
 msgstr ""
 
-#: lib/statsample/regression/multiple/baseengine.rb:161
+#: lib/statsample/regression/multiple/baseengine.rb:175
+#: lib/statsample/regression/multiple/matrixengine.rb:209
 msgid "Regression"
 msgstr ""
 
-#: lib/statsample/regression/multiple/baseengine.rb:162
+#: lib/statsample/regression/multiple/baseengine.rb:176
+#: lib/statsample/regression/multiple/matrixengine.rb:210
 msgid "Error"
 msgstr ""
 
-#: lib/statsample/regression/multiple/baseengine.rb:164
+#: lib/statsample/regression/multiple/baseengine.rb:178
+#: lib/statsample/regression/multiple/matrixengine.rb:212
 #: lib/statsample/crosstab.rb:108 lib/statsample/crosstab.rb:123
 #: lib/statsample/crosstab.rb:159 lib/statsample/crosstab.rb:181
-#: lib/statsample/dominanceanalysis.rb:281
+#: lib/statsample/dominanceanalysis.rb:361
 msgid "Total"
 msgstr ""
 
-#: lib/statsample/regression/multiple/baseengine.rb:170
+#: lib/statsample/regression/multiple/baseengine.rb:184
+#: lib/statsample/regression/multiple/matrixengine.rb:219
+#: lib/statsample/regression/multiple/matrixengine.rb:221
 msgid "Constant"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:42
+#: lib/statsample/regression/multiple/matrixengine.rb:75
+msgid "Multiple reggresion of %s on %s"
+msgstr ""
+
+#: lib/statsample/regression/multiple/matrixengine.rb:202
+msgid "Cases=%d"
+msgstr ""
+
+#: lib/statsample/dominanceanalysis/bootstrap.rb:117
 msgid "Bootstrap dominance Analysis:  %s over %s"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:65
+#: lib/statsample/dominanceanalysis/bootstrap.rb:140
 msgid "Bootstrap %d of %d"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:106
+#: lib/statsample/dominanceanalysis/bootstrap.rb:183
 msgid "DAB: "
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:109
+#: lib/statsample/dominanceanalysis/bootstrap.rb:186
 msgid "Sample size: %d\n"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:111
+#: lib/statsample/dominanceanalysis/bootstrap.rb:188
 msgid "Linear Regression Engine: %s"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:113
+#: lib/statsample/dominanceanalysis/bootstrap.rb:190
 msgid "pairs"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:113
+#: lib/statsample/dominanceanalysis/bootstrap.rb:190
 msgid "SE(Dij)"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:113
+#: lib/statsample/dominanceanalysis/bootstrap.rb:190
 msgid "Reproducibility"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:114
+#: lib/statsample/dominanceanalysis/bootstrap.rb:191
 msgid "Complete dominance"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:122
+#: lib/statsample/dominanceanalysis/bootstrap.rb:199
 msgid "Conditional dominance"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:131
+#: lib/statsample/dominanceanalysis/bootstrap.rb:208
 msgid "General Dominance"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:140
+#: lib/statsample/dominanceanalysis/bootstrap.rb:217
 msgid "General averages"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:140
+#: lib/statsample/dominanceanalysis/bootstrap.rb:217
 msgid "var"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:140
+#: lib/statsample/dominanceanalysis/bootstrap.rb:217
 msgid "mean"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:140
+#: lib/statsample/dominanceanalysis/bootstrap.rb:217
 msgid "se"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:140
+#: lib/statsample/dominanceanalysis/bootstrap.rb:217
 msgid "p.5"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis/bootstrap.rb:140
+#: lib/statsample/dominanceanalysis/bootstrap.rb:217
 msgid "p.95"
 msgstr ""
 
-#: lib/statsample/bivariate/polychoric.rb:560
-#: lib/statsample/bivariate/tetrachoric.rb:118
+#: lib/statsample/bivariate/polychoric.rb:750
+#: lib/statsample/bivariate/tetrachoric.rb:124
 msgid "Contingence Table"
 msgstr ""
 
-#: lib/statsample/bivariate/polychoric.rb:569
+#: lib/statsample/bivariate/polychoric.rb:759
 msgid "Thresholds"
 msgstr ""
 
-#: lib/statsample/bivariate/tetrachoric.rb:126
+#: lib/statsample/bivariate/polychoric.rb:767
+msgid "Test of bivariate normality: X2 = %0.3f, df = %d, p= %0.5f"
+msgstr ""
+
+#: lib/statsample/bivariate/tetrachoric.rb:132
 msgid "SE: %0.3f"
 msgstr ""
 
-#: lib/statsample/bivariate/tetrachoric.rb:127
+#: lib/statsample/bivariate/tetrachoric.rb:133
 msgid "Threshold X: %0.3f "
 msgstr ""
 
-#: lib/statsample/bivariate/tetrachoric.rb:128
+#: lib/statsample/bivariate/tetrachoric.rb:134
 msgid "Threshold Y: %0.3f "
 msgstr ""
 
-#: lib/statsample/bivariate/tetrachoric.rb:134
+#: lib/statsample/bivariate/tetrachoric.rb:141
 msgid "Tetrachoric correlation"
 msgstr ""
 
@@ -152,31 +172,31 @@ msgstr ""
 msgid "Histogram %s"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:99
+#: lib/statsample/factor/pca.rb:128
 msgid "PCA: "
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:100
+#: lib/statsample/factor/pca.rb:129
 msgid "PCA"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:103 lib/statsample/factor/principalaxis.rb:106
+#: lib/statsample/factor/pca.rb:132 lib/statsample/factor/principalaxis.rb:182
 msgid "Communalities"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:109 lib/statsample/factor/principalaxis.rb:112
+#: lib/statsample/factor/pca.rb:138 lib/statsample/factor/principalaxis.rb:188
 msgid "Eigenvalues"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:115 lib/statsample/factor/principalaxis.rb:118
+#: lib/statsample/factor/pca.rb:144 lib/statsample/factor/principalaxis.rb:194
 msgid "Component Matrix"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:97
+#: lib/statsample/factor/principalaxis.rb:176
 msgid "Factor Analysis: "
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:98
+#: lib/statsample/factor/principalaxis.rb:177
 msgid "Factor Analysis"
 msgstr ""
 
@@ -216,50 +236,50 @@ msgstr ""
 msgid "% Total"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:79
+#: lib/statsample/dominanceanalysis.rb:123
 msgid "Dominance Analysis:  %s over %s"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:241
+#: lib/statsample/dominanceanalysis.rb:319
 msgid "DA: "
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:243
-msgid "Dominance Analysis result"
+#: lib/statsample/dominanceanalysis.rb:322
+msgid "sign"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:244
-msgid "sign"
+#: lib/statsample/dominanceanalysis.rb:323
+msgid "Dominance Analysis result"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:245
+#: lib/statsample/dominanceanalysis.rb:325
 msgid "Model 0"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:259
+#: lib/statsample/dominanceanalysis.rb:340
 msgid "k=%d Average"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:272
+#: lib/statsample/dominanceanalysis.rb:352
 msgid "Overall averages"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:281
+#: lib/statsample/dominanceanalysis.rb:361
 msgid "Pairwise dominance"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:281
+#: lib/statsample/dominanceanalysis.rb:361
 msgid "Pairs"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:281
+#: lib/statsample/dominanceanalysis.rb:361
 msgid "Conditional"
 msgstr ""
 
-#: lib/statsample/dominanceanalysis.rb:281
+#: lib/statsample/dominanceanalysis.rb:361
 msgid "General"
 msgstr ""
 
-#: lib/statsample/converters.rb:317
+#: lib/statsample/converters.rb:321
 msgid ")}"
 msgstr ""

data/test/test_anova.rb

@@ -1,8 +1,8 @@
 $:.unshift(File.dirname(__FILE__)+'/../lib/')
 require 'statsample'
-require 'test/unit'
-
-class StatsampleAnovaTestCase < Test::Unit::TestCase
+require 'minitest/unit'
+MiniTest::Unit.autorun
+class StatsampleAnovaTestCase < MiniTest::Unit::TestCase
 	def initialize(*args)
     @v1=[3,3,2,3,6].to_vector(:scale)
     @v2=[7,6,5,6,7].to_vector(:scale)

data/test/test_bivariate.rb

@@ -1,19 +1,47 @@
 $:.unshift(File.dirname(__FILE__)+'/../lib/')
 require 'statsample'
-require 'test/unit'
-class StatsampleBivariateTestCase < Test::Unit::TestCase
-  def test_sum_of_codeviated
+require 'minitest/unit'
+MiniTest::Unit.autorun
+class StatsampleBivariateTestCase < MiniTest::Unit::TestCase
+  def test_sum_of_squares
     v1=[1,2,3,4,5,6].to_vector(:scale)
     v2=[6,2,4,10,12,8].to_vector(:scale)
-    assert_equal(23.0, Statsample::Bivariate.sum_of_codeviated(v1,v2))
+    assert_equal(23.0, Statsample::Bivariate.sum_of_squares(v1,v2))
+  end
+  def test_covariance
+    if Statsample.has_gsl?
+      v1=1000.times.collect {|a| rand()}.to_scale
+      v2=1000.times.collect {|a| rand()}.to_scale
+      assert_in_delta(Statsample::Bivariate.covariance(v1,v2), Statsample::Bivariate.covariance_slow(v1,v2), 0.001)
+    else
+      puts "Bivariate::covariance not tested (needs GSL)"
+    end
+  
+  end
+
+  def test_gsl_pearson
+     if Statsample.has_gsl?
+       v1=100.times.collect {|a| rand()}.to_scale
+       v2=100.times.collect {|a| rand()}.to_scale
+
+       assert_in_delta(GSL::Stats::correlation(v1.gsl, v2.gsl), Statsample::Bivariate.pearson_slow(v1,v2), 1e-10)
+     else
+       puts "Not tested gsl versus ruby correlation (needs GSL)"
+     end
   end
   def test_pearson
     v1=[6,5,4,7,8,4,3,2].to_vector(:scale)
     v2=[2,3,7,8,6,4,3,2].to_vector(:scale)
     assert_in_delta(0.525,Statsample::Bivariate.pearson(v1,v2), 0.001)
+    assert_in_delta(0.525,Statsample::Bivariate.pearson_slow(v1,v2), 0.001)
+    
     v3=[6,2,  1000,1000,5,4,7,8,4,3,2,nil].to_vector(:scale)
     v4=[2,nil,nil,nil,  3,7,8,6,4,3,2,500].to_vector(:scale)
     assert_in_delta(0.525,Statsample::Bivariate.pearson(v3,v4),0.001)
+    # Test ruby method
+    v3a,v4a=Statsample.only_valid v3, v4
+    assert_in_delta(0.525, Statsample::Bivariate.pearson_slow(v3a,v4a),0.001)
+
   end
   def test_tetrachoric_matrix
     ds=Statsample::PlainText.read(File.dirname(__FILE__)+"/../data/tetmat_test.txt", %w{a b c d e})
@@ -26,7 +54,7 @@ class StatsampleBivariateTestCase < Test::Unit::TestCase
     end
   end
   def test_poly_vs_tetra
-    10.times {
+    5.times {
       # Should be the same results as Tetrachoric for 2x2 matrix
       matrix=Matrix[[150+rand(10),1000+rand(20)],[1000+rand(20),200+rand(20)]]
       tetra = Statsample::Bivariate::Tetrachoric.new_with_matrix(matrix)
@@ -49,16 +77,6 @@ class StatsampleBivariateTestCase < Test::Unit::TestCase
       assert_in_delta(-0.027, poly.threshold_x[0],0.001)
       assert_in_delta(1.578, poly.threshold_y[1],0.001)
       assert_in_delta(1.137, poly.threshold_x[1],0.001)
-      
-      
-
-      
-      
-     
-      
-      
-
-      
     if Statsample.has_gsl?  
       poly.method=:polychoric_series
       poly.compute
@@ -95,17 +113,18 @@ class StatsampleBivariateTestCase < Test::Unit::TestCase
     end
     assert(poly.summary)
   end
+  
   def test_tetrachoric
     a,b,c,d=0,0,0,0
-    assert_raise RuntimeError do
+    assert_raises RuntimeError do
       tc  = Statsample::Bivariate::Tetrachoric.new(a,b,c,d)
     end
     a,b,c,d=10,10,0,0
-    assert_raise RuntimeError do
+    assert_raises RuntimeError do
       tc  = Statsample::Bivariate::Tetrachoric.new(a,b,c,d)
     end
     a,b,c,d=10,0,10,0
-    assert_raise RuntimeError do
+    assert_raises RuntimeError do
       tc  = Statsample::Bivariate::Tetrachoric.new(a,b,c,d)
     end
     a,b,c,d=10,0,0,10
@@ -174,14 +193,6 @@ class StatsampleBivariateTestCase < Test::Unit::TestCase
     assert(Statsample::Bivariate.prop_pearson(t,n,:right)>0.05)
     assert(Statsample::Bivariate.prop_pearson(t,n,:left)<0.05)
   end
-  def test_covariance
-  if Statsample.has_gsl?
-    v1=[6,5,4,7,8,4,3,2].to_vector(:scale)
-    v2=[2,3,7,8,6,4,3,2].to_vector(:scale)
-    assert_in_delta(Statsample::Bivariate.covariance(v1,v2), Statsample::Bivariate.covariance_slow(v1,v2), 0.001)
-    
-  end
-  end
   
   def test_spearman
     v1=[86,97,99,100,101,103,106,110,112,113].to_vector(:scale)
@@ -192,7 +203,7 @@ class StatsampleBivariateTestCase < Test::Unit::TestCase
   def test_point_biserial
     c=[1,3,5,6,7,100,200,300,400,300].to_vector(:scale)
     d=[1,1,1,1,1,0,0,0,0,0].to_vector(:scale)
-    assert_raise TypeError do
+    assert_raises TypeError do
       Statsample::Bivariate.point_biserial(c,d)
     end
     assert_in_delta(Statsample::Bivariate.point_biserial(d,c), Statsample::Bivariate.pearson(d,c), 0.0001)

data/test/test_csv.rb

@@ -1,15 +1,15 @@
 $:.unshift(File.dirname(__FILE__)+'/../lib/')
 require 'statsample'
 require "tempfile"
-require 'test/unit'
-
-class StatsampleCSVTestCase < Test::Unit::TestCase
+require 'minitest/unit'
+MiniTest::Unit.autorun
+class StatsampleCSVTestCase < MiniTest::Unit::TestCase
     def setup
         @ds=Statsample::CSV.read(File.dirname(__FILE__)+"/test_csv.csv")
     end
     def test_read
         assert_equal(6,@ds.cases)
-        assert_equal(%w{id name age city a1},@ds.fields)
+        assert_equal(%w{id name age city a1}, @ds.fields)
         id=[1,2,3,4,5,6].to_vector(:scale)
         name=["Alex","Claude","Peter","Franz","George","Fernand"].to_vector(:nominal)
         age=[20,23,25,27,5.5,nil].to_vector(:scale)