statsample 0.14.0 → 0.14.1

data/po/statsample.pot

@@ -6,8 +6,8 @@
 #, fuzzy
 msgid ""
 msgstr ""
-"Project-Id-Version: statsample 0.13.1\n"
-"POT-Creation-Date: 2010-08-10 17:00-0400\n"
+"Project-Id-Version: statsample 0.14.0\n"
+"POT-Creation-Date: 2010-08-17 18:46-0400\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"
@@ -24,8 +24,8 @@ msgstr ""
 msgid "Mean and standard deviation"
 msgstr ""
 
-#: lib/statsample/test/t.rb:209 lib/statsample/factor/pca.rb:169
-#: lib/statsample/factor/principalaxis.rb:190
+#: lib/statsample/test/t.rb:209 lib/statsample/factor/pca.rb:163
+#: lib/statsample/factor/principalaxis.rb:202
 msgid "Variable"
 msgstr ""
 
@@ -38,7 +38,8 @@ msgstr ""
 msgid "sd"
 msgstr ""
 
-#: lib/statsample/test/t.rb:209 lib/statsample/factor/parallelanalysis.rb:79
+#: lib/statsample/test/t.rb:209 lib/statsample/factor/parallelanalysis.rb:100
+#: lib/statsample/factor/parallelanalysis.rb:108
 msgid "n"
 msgstr ""
 
@@ -91,6 +92,10 @@ msgstr ""
 msgid "Levene Test"
 msgstr ""
 
+#: lib/statsample/test/bartlettsphericity.rb:25
+msgid "Bartlett's test of sphericity"
+msgstr ""
+
 #: lib/statsample/regression/multiple/baseengine.rb:25
 msgid "Multiple Regression:  %s over %s"
 msgstr ""
@@ -266,11 +271,11 @@ msgstr ""
 msgid "Descriptives"
 msgstr ""
 
-#: lib/statsample/bivariate/pearson.rb:33
+#: lib/statsample/bivariate/pearson.rb:32
 msgid "Correlation (%s - %s)"
 msgstr ""
 
-#: lib/statsample/bivariate/pearson.rb:51
+#: lib/statsample/bivariate/pearson.rb:50
 msgid "%s : r=%0.3f (t:%0.3f, g.l.=%d, p:%0.3f / %s tails)"
 msgstr ""
 
@@ -278,143 +283,186 @@ msgstr ""
 msgid "Histogram %s"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:51
+#: lib/statsample/factor/parallelanalysis.rb:70
 msgid "Parallel Analysis"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:73
+#: lib/statsample/factor/parallelanalysis.rb:93
 msgid "Bootstrap Method: %s"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:74
+#: lib/statsample/factor/parallelanalysis.rb:94
+msgid "Uses SMC: %s"
+msgstr ""
+
+#: lib/statsample/factor/parallelanalysis.rb:94
+msgid "Yes"
+msgstr ""
+
+#: lib/statsample/factor/parallelanalysis.rb:94
+msgid "No"
+msgstr ""
+
+#: lib/statsample/factor/parallelanalysis.rb:95
 msgid "Correlation Matrix type : %s"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:75
+#: lib/statsample/factor/parallelanalysis.rb:96
 msgid "Number of variables: %d"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:76
+#: lib/statsample/factor/parallelanalysis.rb:97
 msgid "Number of cases: %d"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:77
+#: lib/statsample/factor/parallelanalysis.rb:98
 msgid "Number of iterations: %d"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:78
-msgid "Number or factors to preserve: %d"
+#: lib/statsample/factor/parallelanalysis.rb:100
+#: lib/statsample/factor/parallelanalysis.rb:108
+#: lib/statsample/factor/map.rb:90
+msgid "Eigenvalues"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:79
-msgid "Eigenvalues"
+#: lib/statsample/factor/parallelanalysis.rb:100
+#: lib/statsample/factor/parallelanalysis.rb:108
+msgid "generated eigenvalue"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:79
-msgid "data eigenvalue"
+#: lib/statsample/factor/parallelanalysis.rb:107
+msgid "Number or factors to preserve: %d"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:79
-msgid "generated eigenvalue"
+#: lib/statsample/factor/parallelanalysis.rb:108
+msgid "data eigenvalue"
 msgstr ""
 
-#: lib/statsample/factor/parallelanalysis.rb:79
+#: lib/statsample/factor/parallelanalysis.rb:108
 msgid "preserve?"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:51
+#: lib/statsample/factor/map.rb:54
+msgid "Velicer's MAP"
+msgstr ""
+
+#: lib/statsample/factor/map.rb:90
+msgid "Value"
+msgstr ""
+
+#: lib/statsample/factor/map.rb:95
+msgid "Velicer's Average Squared Correlations"
+msgstr ""
+
+#: lib/statsample/factor/map.rb:95
+msgid "number of components"
+msgstr ""
+
+#: lib/statsample/factor/map.rb:95
+msgid "average square correlation"
+msgstr ""
+
+#: lib/statsample/factor/map.rb:100
+msgid "The smallest average squared correlation is : %0.6f"
+msgstr ""
+
+#: lib/statsample/factor/map.rb:101
+msgid "The number of components is : %d"
+msgstr ""
+
+#: lib/statsample/factor/pca.rb:50
 msgid "Principal Component Analysis"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:118
+#: lib/statsample/factor/pca.rb:117
 msgid "Component matrix"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:168 lib/statsample/factor/principalaxis.rb:188
+#: lib/statsample/factor/pca.rb:162 lib/statsample/factor/principalaxis.rb:200
 msgid "Number of factors: %d"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:169 lib/statsample/factor/principalaxis.rb:190
+#: lib/statsample/factor/pca.rb:163 lib/statsample/factor/principalaxis.rb:202
 msgid "Communalities"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:169 lib/statsample/factor/principalaxis.rb:190
+#: lib/statsample/factor/pca.rb:163 lib/statsample/factor/principalaxis.rb:202
 msgid "Initial"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:169 lib/statsample/factor/principalaxis.rb:190
+#: lib/statsample/factor/pca.rb:163 lib/statsample/factor/principalaxis.rb:202
 msgid "Extraction"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:175
+#: lib/statsample/factor/pca.rb:169
 msgid "Total Variance Explained"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:175
+#: lib/statsample/factor/pca.rb:169
 msgid "Component"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:175
+#: lib/statsample/factor/pca.rb:169
 msgid "E.Total"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:175
+#: lib/statsample/factor/pca.rb:169
 msgid "%"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:175
+#: lib/statsample/factor/pca.rb:169
 msgid "Cum. %"
 msgstr ""
 
-#: lib/statsample/factor/pca.rb:179
+#: lib/statsample/factor/pca.rb:173
 msgid "Component %d"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:64
+#: lib/statsample/factor/principalaxis.rb:71
 msgid "Variable %d"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:140
+#: lib/statsample/factor/principalaxis.rb:147
 msgid "Factor Matrix"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:189
+#: lib/statsample/factor/principalaxis.rb:201
 msgid "Iterations: %d"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:195
+#: lib/statsample/factor/principalaxis.rb:207
 msgid "Total Variance"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:195
+#: lib/statsample/factor/principalaxis.rb:207
 msgid "Factor"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:195
+#: lib/statsample/factor/principalaxis.rb:207
 msgid "I.E.Total"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:195
+#: lib/statsample/factor/principalaxis.rb:207
 msgid "I.E. %"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:195
+#: lib/statsample/factor/principalaxis.rb:207
 msgid "I.E.Cum. %"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:196
+#: lib/statsample/factor/principalaxis.rb:208
 msgid "S.L.Total"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:196
+#: lib/statsample/factor/principalaxis.rb:208
 msgid "S.L. %"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:196
+#: lib/statsample/factor/principalaxis.rb:208
 msgid "S.L.Cum. %"
 msgstr ""
 
-#: lib/statsample/factor/principalaxis.rb:203
+#: lib/statsample/factor/principalaxis.rb:215
 msgid "Factor %d"
 msgstr ""
 
@@ -430,31 +478,31 @@ msgstr ""
 msgid "Component transformation matrix"
 msgstr ""
 
-#: lib/statsample/reliability/multiscaleanalysis.rb:57
+#: lib/statsample/reliability/multiscaleanalysis.rb:58
 msgid "Multiple Scale analysis"
 msgstr ""
 
-#: lib/statsample/reliability/multiscaleanalysis.rb:85
+#: lib/statsample/reliability/multiscaleanalysis.rb:86
 msgid "Scale %s"
 msgstr ""
 
-#: lib/statsample/reliability/multiscaleanalysis.rb:125
+#: lib/statsample/reliability/multiscaleanalysis.rb:128
 msgid "Reliability analysis of scales"
 msgstr ""
 
-#: lib/statsample/reliability/multiscaleanalysis.rb:131
+#: lib/statsample/reliability/multiscaleanalysis.rb:134
 msgid "Correlation matrix for %s"
 msgstr ""
 
-#: lib/statsample/reliability/multiscaleanalysis.rb:136
+#: lib/statsample/reliability/multiscaleanalysis.rb:139
 msgid "PCA for %s"
 msgstr ""
 
-#: lib/statsample/reliability/multiscaleanalysis.rb:141
+#: lib/statsample/reliability/multiscaleanalysis.rb:144
 msgid "Principal Axis for %s"
 msgstr ""
 
-#: lib/statsample/reliability/multiscaleanalysis.rb:147
+#: lib/statsample/reliability/multiscaleanalysis.rb:150
 msgid "Parallel Analysis for %s"
 msgstr ""
 

data/test/fixtures/correlation_matrix.rb

@@ -0,0 +1,17 @@
+# Retrieve Correlation matrix for eigth variables
+module Statsample
+  module Fixtures
+    def harman_817
+      Matrix[
+    [1.0,   0.84,   0.62,   -0.53,   0.03,    0.57,   -0.33,  -0.63],
+    [0.84,  1.00,   0.84,   -0.68,  -0.05,    0.76,   -0.35,  -0.73],
+    [0.62,  0.84,   1.00,   -0.76,   0.08,    0.81,   -0.51,  -0.81],
+    [-0.53, -0.68,  -0.76,  1.00,    -0.25,  -0.80,    0.62,   0.88],
+    [0.03,  -0.05,  0.08,   -0.25,   1.00,    0.25,   -0.72,  -0.36],
+    [0.57,   0.76,  0.81,   -0.80,   0.25,    1.00,   -0.58,  -0.84],
+    [-0.33, -0.35, -0.51,    0.62,  -0.72,   -0.58,    1.00,   0.68],
+    [-0.63, -0.73, -0.81,    0.88,  -0.36,   -0.84,    0.68,   1.00]
+    ].extend(Statsample::CovariateMatrix)
+    end
+  end
+end

data/test/helpers_tests.rb

@@ -1,10 +1,12 @@
 $:.unshift(File.dirname(__FILE__)+'/../lib/')
+$:.unshift(File.dirname(__FILE__)+'/')
 
 require 'statsample'
 require 'minitest/unit'
 require 'tempfile'
 require 'tmpdir'
 require 'shoulda'
+require 'fixtures/correlation_matrix'
 
 module MiniTest
   class Unit

data/test/test_bartlettsphericity.rb

@@ -0,0 +1,26 @@
+require(File.dirname(__FILE__)+'/helpers_tests.rb')
+
+
+class StatsampleBartlettSphericityTestCase < MiniTest::Unit::TestCase
+  include Statsample::Test
+  context Statsample::Test::BartlettSphericity do
+    setup do
+      @v1=[1 ,2 ,3 ,4 ,7 ,8 ,9 ,10,14,15,20,50,60,70].to_scale
+      @v2=[5 ,6 ,11,12,13,16,17,18,19,20,30,0,0,0].to_scale
+      @v3=[10,3 ,20,30,40,50,80,10,20,30,40,2,3,4].to_scale
+      # KMO: 0.490
+      ds={'v1'=>@v1,'v2'=>@v2,'v3'=>@v3}.to_dataset
+      cor=Statsample::Bivariate.correlation_matrix(ds)
+      @bs=Statsample::Test::BartlettSphericity.new(cor, 14)
+    end
+    should "have correct value for chi" do
+      assert_in_delta(9.477, @bs.value,0.001)
+    end
+    should "have correct value for df" do
+      assert_equal(3, @bs.df)
+    end
+    should "have correct value for probability" do
+      assert_in_delta(0.024,@bs.probability,0.001)
+    end
+  end
+end

data/test/test_factor.rb

@@ -1,11 +1,70 @@
 require(File.dirname(__FILE__)+'/helpers_tests.rb')
 
 class StatsampleFactorTestCase < MiniTest::Unit::TestCase
+  include Statsample::Fixtures
+  def test_antiimage
+    cor=Matrix[[1,0.964, 0.312],[0.964,1,0.411],[0.312,0.411,1]]
+    expected=Matrix[[0.062,-0.057, 0.074],[-0.057, 0.057, -0.089], [0.074, -0.089, 0.729]]
+    ai=Statsample::Factor.anti_image_covariance_matrix(cor)
+    assert(Matrix.equal_in_delta?(expected, ai, 0.01), "#{expected.to_s} not equal to #{ai.to_s}")
+  end
+  def test_kmo
+      @v1=[1 ,2 ,3 ,4 ,7 ,8 ,9 ,10,14,15,20,50,60,70].to_scale
+      @v2=[5 ,6 ,11,12,13,16,17,18,19,20,30,0,0,0].to_scale
+      @v3=[10,3 ,20,30,40,50,80,10,20,30,40,2,3,4].to_scale
+      # KMO: 0.490
+      ds={'v1'=>@v1,'v2'=>@v2,'v3'=>@v3}.to_dataset
+      cor=Statsample::Bivariate.correlation_matrix(ds)
+     kmo=Statsample::Factor.kmo(cor)
+     assert_in_delta(0.667, kmo,0.001)
+     assert_in_delta(0.81, Statsample::Factor.kmo(harman_817),0.01)
+     
+  end
+  def test_kmo_univariate
+    m=harman_817
+    expected=[0.73,0.76,0.84,0.87,0.53,0.93,0.78,0.86]
+    m.row_size.times.map {|i|
+      assert_in_delta(expected[i], Statsample::Factor.kmo_univariate(m,i),0.01)
+    }
+  end
+  def test_parallelanalysis_with_data
+    samples=100
+    variables=10
+    iterations=50
+    rng = GSL::Rng.alloc()
+    f1=samples.times.collect {rng.ugaussian()}.to_scale
+    f2=samples.times.collect {rng.ugaussian()}.to_scale    
+    vectors={}
+    variables.times do |i|
+      if i<5
+        vectors["v#{i}"]=samples.times.collect {|nv|
+          f1[nv]*5+f2[nv]*2+rng.ugaussian()
+        }.to_scale
+      else
+        vectors["v#{i}"]=samples.times.collect {|nv|
+          f2[nv]*5+f1[nv]*2+rng.ugaussian()
+        }.to_scale
+      end
+      
+    end
+    ds=vectors.to_dataset
+    
+    pa1=Statsample::Factor::ParallelAnalysis.new(ds, :bootstrap_method=>:data, :iterations=>iterations)
+    pa2=Statsample::Factor::ParallelAnalysis.with_random_data(samples,variables,:iterations=>iterations,:percentil=>95)
+    3.times do |n|
+      var="ev_0000#{n+1}"
+      assert_in_delta(pa1.ds_eigenvalues[var].mean,pa2.ds_eigenvalues[var].mean,0.04)
+    end
+  end
   def test_parallelanalysis
-    pa=Statsample::Factor::ParallelAnalysis.with_random_data(305,8,150)
+    pa=Statsample::Factor::ParallelAnalysis.with_random_data(305,8,:iterations=>100,:percentil=>95)
     assert_in_delta(1.2454, pa.ds_eigenvalues['ev_00001'].mean, 0.01)
     assert_in_delta(1.1542, pa.ds_eigenvalues['ev_00002'].mean, 0.01)
     assert_in_delta(1.0836, pa.ds_eigenvalues['ev_00003'].mean, 0.01)
+    #puts pa.summary
+    assert(pa.summary.size>0)
+    #pa=Statsample::Factor::ParallelAnalysis.with_random_data(305,8,100, 95, true)
+    #puts pa.summary
   end
   def test_map
     fields=%w{height arm.span forearm lower.leg weight bitro.diameter chest.girth chest.width}
@@ -69,8 +128,10 @@ class StatsampleFactorTestCase < MiniTest::Unit::TestCase
   def test_principalaxis
       matrix=::Matrix[
       [1.0, 0.709501601093587, 0.877596585880047, 0.272219316266807],  [0.709501601093587, 1.0, 0.291633797330304, 0.871141831433844], [0.877596585880047, 0.291633797330304, 1.0, -0.213373722977167], [0.272219316266807, 0.871141831433844, -0.213373722977167, 1.0]]
-      fa=Statsample::Factor::PrincipalAxis.new(matrix,:m=>1, :max_iterations=>50)
       
+      
+      fa=Statsample::Factor::PrincipalAxis.new(matrix,:m=>1, :max_iterations=>50)
+
       cm=::Matrix[[0.923],[0.912],[0.507],[0.483]]
       
       _test_matrix(cm,fa.component_matrix)