statsample 0.6.2 → 0.6.3

data/test/test_bivariate.rb

@@ -33,14 +33,37 @@ class StatsampleBivariateTestCase < Test::Unit::TestCase
     poly  = Statsample::Bivariate::Polychoric.new(matrix)
     assert_in_delta(tetra.r,poly.r,0.0001)
     
-    # Example for http://www.john-uebersax.com/stat/tetra.htm#exampl
+    # Example for Tallis(1962, cited by Drasgow, 2006)
     
     matrix=Matrix[[58,52,1],[26,58,3],[8,12,9]]
     poly=Statsample::Bivariate::Polychoric.new(matrix)
-    assert_in_delta(0.4199, poly.r, 0.0001)
-    assert_in_delta(-0.2397, poly.threshold_y[0],0.001)
-    assert_in_delta(-0.0276, poly.threshold_x[0],0.001)
+    poly.method=:two_step
+    poly.compute
+    assert_in_delta(0.420, poly.r, 0.001)
+    assert_in_delta(-0.240, poly.threshold_y[0],0.001)
+    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)
     
+    
+    poly.method=:polychoric_series
+    poly.compute
+    
+    assert_in_delta(0.556, poly.r, 0.001)
+    assert_in_delta(-0.240, poly.threshold_y[0],0.001)
+    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)
+    
+    
+    poly.method=:joint
+    poly.compute
+    
+    assert_in_delta(0.4192, poly.r, 0.0001)
+    assert_in_delta(-0.2421, poly.threshold_y[0],0.0001)
+    assert_in_delta(-0.0297, poly.threshold_x[0],0.0001)
+    assert_in_delta(1.5938, poly.threshold_y[1],0.0001)
+    assert_in_delta(1.1331, poly.threshold_x[1],0.0001)
 
   end
   def test_tetrachoric

data/test/test_distribution.rb

@@ -10,34 +10,34 @@ end
 class DistributionTestCase < Test::Unit::TestCase
   def test_chi
     if !NOT_GSL
-    [2,3,4,5].each{|k|
+      [2,3,4,5].each{|k|
         chis=rand()*10
         area=Distribution::ChiSquare.cdf(chis, k)
         assert_in_delta(area, GSL::Cdf.chisq_P(chis,k),0.0001)
         assert_in_delta(chis, Distribution::ChiSquare.p_value(area,k),0.0001,"Error on prob #{area} and k #{k}")
-    }
+      }
     end
   end
   def test_t
     if !NOT_GSL
-        [-2,0.1,0.5,1,2].each{|t|
-            [2,5,10].each{|n|
-                area=Distribution::T.cdf(t,n)
-                assert_in_delta(area, GSL::Cdf.tdist_P(t,n),0.0001)
-                assert_in_delta(Distribution::T.p_value(area,n), GSL::Cdf.tdist_Pinv(area,n),0.0001)
-                
-            }
+      [-2,0.1,0.5,1,2].each{|t|
+        [2,5,10].each{|n|
+          area=Distribution::T.cdf(t,n)
+          assert_in_delta(area, GSL::Cdf.tdist_P(t,n),0.0001)
+          assert_in_delta(Distribution::T.p_value(area,n), GSL::Cdf.tdist_Pinv(area,n),0.0001)
+          
         }
+      }
     end
   end
   def test_normal
     if !NOT_GSL
-        [-2,0.1,0.5,1,2].each{|x|
-            area=Distribution::Normal.cdf(x)
-            assert_in_delta(area, GSL::Cdf.ugaussian_P(x),0.0001)
-            assert_in_delta(Distribution::Normal.p_value(area), GSL::Cdf.ugaussian_Pinv(area),0.0001)
-            assert_in_delta(Distribution::Normal.pdf(x), GSL::Ran::ugaussian_pdf(x),0.0001)
-        }
+      [-2,0.1,0.5,1,2].each{|x|
+        area=Distribution::Normal.cdf(x)
+        assert_in_delta(area, GSL::Cdf.ugaussian_P(x),0.0001)
+        assert_in_delta(Distribution::Normal.p_value(area), GSL::Cdf.ugaussian_Pinv(area),0.0001)
+        assert_in_delta(Distribution::Normal.pdf(x), GSL::Ran::ugaussian_pdf(x),0.0001)
+      }
     end
   end
   def test_normal_bivariate
@@ -49,7 +49,8 @@ class DistributionTestCase < Test::Unit::TestCase
     end
     
     [-3,-2,-1,0,1,1.5].each {|x|
-      assert_in_delta(Distribution::NormalBivariate.cdf_math(x,x,0.5), Distribution::NormalBivariate.cdf_iterate(x,x,0.5), 0.001)
+      assert_in_delta(Distribution::NormalBivariate.cdf_hull(x,x,0.5), Distribution::NormalBivariate.cdf_genz(x,x,0.5), 0.001)
+      assert_in_delta(Distribution::NormalBivariate.cdf_genz(x,x,0.5), Distribution::NormalBivariate.cdf_jantaravareerat(x,x,0.5), 0.001)
     }
     
     assert_in_delta(0.686, Distribution::NormalBivariate.cdf(2,0.5,0.5), 0.001)

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: statsample
 version: !ruby/object:Gem::Version 
-  version: 0.6.2
+  version: 0.6.3
 platform: ruby
 authors: 
 - Claudio Bustos
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-02-11 00:00:00 -03:00
+date: 2010-02-15 00:00:00 -03:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -52,7 +52,7 @@ dependencies:
       - !ruby/object:Gem::Version 
         version: 2.4.0
     version: 
-description: A suite for your basic and advanced statistics needs. Descriptive statistics, multiple regression, factorial analysis, dominance analysis, scale's reliability analysis, bivariate statistics and others procedures.
+description: ""
 email: 
 - clbustos@gmail.com
 executables: 
@@ -76,6 +76,7 @@ files:
 - data/test_binomial.csv
 - data/tetmat_matrix.txt
 - data/tetmat_test.txt
+- demo/correlation_matrix.rb
 - demo/dominance_analysis_bootstrap.rb
 - demo/dominanceanalysis.rb
 - demo/multiple_regression.rb
@@ -114,7 +115,6 @@ files:
 - lib/statsample/graph/svghistogram.rb
 - lib/statsample/graph/svgscatterplot.rb
 - lib/statsample/histogram.rb
-- lib/statsample/htmlreport.rb
 - lib/statsample/matrix.rb
 - lib/statsample/mle.rb
 - lib/statsample/mle/logit.rb
@@ -197,7 +197,7 @@ rubyforge_project: ruby-statsample
 rubygems_version: 1.3.5
 signing_key: 
 specification_version: 3
-summary: A suite for your basic and advanced statistics needs
+summary: ""
 test_files: 
 - test/test_bivariate.rb
 - test/test_factor.rb

data/lib/statsample/htmlreport.rb

@@ -1,255 +0,0 @@
-require 'statsample/graph/svggraph'
-
-module Statsample
-    class HtmlReport
-    def initialize(name,dir=nil)
-        require 'fileutils'
-        @uniq=1
-        @uniq_file=0
-        @name=name
-        @partials=[]
-        @anchors=[]
-        dir||=@name+"/"
-        @dir=dir
-        @level=1
-        FileUtils.mkdir(@dir) if !File.exists? @dir
-    end
-    def add_summary(name,summary)
-        add_anchor(name)
-        @partials.push(summary)
-    end
-    def add_anchor(name)
-        @anchors.push([name,@level,@uniq])
-        @partials.push("<a name='#{@uniq}'> </a>")
-        @uniq+=1
-    end
-    def uniq_file(prepend="file")
-        @uniq_file+=1
-        "#{prepend}_#{@uniq_file}_#{Time.now.to_i}"
-    end
-    
-    def add_tetrachoric_correlation_matrix(ds)
-      add_anchor("Tetrachoric correlation Matrix")
-      html="<h2>Tetrachoric Correlation Matrix</h2> <table><thead><th>-</th><th>"+ds.fields.join("</th><th>")+"</th> </thead> <tbody>"
-        matrix=Statsample::Bivariate.tetrachoric_correlation_matrix(ds)
-
-      
-        (0...(matrix.row_size)).each {|row|
-            html+="<tr><td>"+ds.fields[row]+"</td>"
-            (0...(matrix.column_size)).each {|col|
-                if matrix[row,col].nil?
-                    html+="<td>--</td>"
-                else
-                    html+="<td><strong>#{sprintf("%0.2f",matrix[row,col])}</td>"
-                end
-            }
-            html+="</tr>"
-        }
-        html+="</tbody></table>"
-        @partials.push(html)
-    end
-    
-    
-    def add_correlation_matrix(ds)
-        add_anchor("Correlation Matrix")
-        html="<h2>Correlation Matrix</h2> <table><thead><th>-</th><th>"+ds.fields.join("</th><th>")+"</th> </thead> <tbody>"
-        matrix=Statsample::Bivariate.correlation_matrix(ds)
-        pmatrix=Statsample::Bivariate.correlation_probability_matrix(ds)
-
-      
-        (0...(matrix.row_size)).each {|row|
-            html+="<tr><td>"+ds.fields[row]+"</td>"
-            (0...(matrix.column_size)).each {|col|
-                if matrix[row,col].nil?
-                    html+="<td>--</td>"
-                else
-                    sig=""
-                    prob_out=""
-                    if !pmatrix[row,col].nil?
-                        prob=pmatrix[row,col]
-                        prob_out=sprintf("%0.3f",prob)
-                        if prob<0.01
-                            sig="**"
-                        elsif prob<0.05
-                            sig="*"
-                        else
-                            sig=""
-                        end
-                    end
-                    if sig==""
-                        html+="<td>#{sprintf("%0.3f",matrix[row,col])} #{sig}<br /> #{prob_out}</td>"
-                    else
-                        html+="<td><strong>#{sprintf("%0.3f",matrix[row,col])} #{sig}<br /> #{prob_out}</strong></td>"
-
-                    end
-                end
-            }
-            html+="</tr>"
-        }
-        html+="</tbody></table>"
-        @partials.push(html)
-    end
-    # Add a scale
-    # First arg is the name of the scale                          
-    # Other are fields
-    def add_scale(ds,name, fields,icc=false)
-        raise "Fields are empty" if fields.size==0
-        add_anchor("Scale:#{name}")
-        
-        ds_partial=ds.dup(fields)
-        ia=Statsample::Reliability::ItemAnalysis.new(ds_partial)
-        html="<h2>Scale: #{name}</h2>"
-        html << ia.html_summary
-        @partials.push(html)
-        @level+=1
-        v=ds_partial.vector_mean
-            add_histogram(name, v)        
-            add_runsequence_plot(name, v)        
-            add_normalprobability_plot(name,v)
-            add_icc(name,fields) if icc
-        @level-=1
-    end
-    
-    def add_boxplot(name,vector,options={})
-        add_graph("Box Plot #{name}", name, vector.svggraph_boxplot(options))
-    end    
-    def add_graph(name,id,graph)
-        add_anchor(name)
-        rs_file=@dir+"/#{uniq_file()}.svg"
-        html = "<h3>#{name}</h3> <p><embed src='#{rs_file}'  width='#{graph.width}' height='#{graph.height}' type='image/svg+xml' /></p>\n"
-        File.open(rs_file, "w") {|f|
-            f.puts(graph.burn)
-        }
-        @partials.push(html)
-    end
-    def add_runsequence_plot(name, vector,options={})
-        add_graph("Run-Sequence Plot #{name}", name, vector.svggraph_runsequence_plot(options))
-    end
-    def add_lag_plot(name,vector, options={})
-        add_graph("Lag Plot #{name}", name,vector.svggraph_lag_plot(options))
-    end
-        
-    def add_normalprobability_plot(name,vector,options={})
-        add_graph("Normal Probability Plot #{name}", name, vector.svggraph_normalprobability_plot(options))
-    end
-
-    def add_scatterplot(name, ds,x_field=nil, y_fields=nil,config={})
-        add_anchor("Scatterplot: #{name}")
-        x_field||=ds.fields[0]
-        y_fields||=ds.fields-[x_field]
-        ds_partial=ds.dup([x_field]+y_fields)
-        sc=Statsample::Graph::SvgScatterplot.new(ds_partial, config)
-        sc.parse
-        sc_file=@dir+"/#{uniq_file("sc")}.svg"
-        html = "<h3>Scatterplot #{name}</h3> <p><embed src='#{sc_file}'  width='#{sc.width}' height='#{sc.height}' type='image/svg+xml' /></p>\n"
-        File.open(sc_file, "w") {|f|
-              f.puts(sc.burn)
-        }
-        @partials.push(html)
-    end
-    
-    
-    def add_boxplots(name, ds,options={})
-        add_anchor("Boxplots: #{name}")
-        options={:graph_title=>"Boxplots:#{name}", :show_graph_title=>true, :height=>500}.merge! options
-        graph = Statsample::Graph::SvgBoxplot.new(options)
-        ds.fields.each{|f|
-            graph.add_data(:title=>f, 
-                :data=>ds[f].valid_data,
-                :vector=>ds[f]
-                )
-        }
-        add_graph(name,name,graph)
-        graph
-    end
-    def add_histogram(name,vector,bins=nil,options={})
-        bins||=vector.size / 15
-        bins=15 if bins>15 
-        graph=vector.svggraph_histogram(bins,options)
-        add_graph("Histogram:#{name}",name,graph)
-        html = "<ul><li>Skewness=#{sprintf("%0.3f",vector.skew)}</li>
-        <li>Kurtosis=#{sprintf("%0.3f",vector.kurtosis)}</li></ul>"
-        @partials.push(html)
-    end
-    def add_icc(name,ds, fields)
-        require 'statsample/graph/svggraph'
-        raise "Fields are empty" if fields.size==0
-        add_anchor("ICC:#{name}")        
-        ds_partial=ds.dup(fields)
-        ia=Statsample::Reliability::ItemAnalysis.new(ds_partial)
-        html="<h3>ICC for scale: #{name}</h3>"
-        ia.svggraph_item_characteristic_curve(@dir ,name, {:width=>400,:height=>300})
-        ds_partial.fields.sort.each{|f|
-            html << "<div><p><strong>#{f}</strong></p><embed src='#{@dir}/#{name}_#{f}.svg'  width='400' height='300' type='image/svg+xml' /></div>\n"
-        }
-        @partials.push(html)
-    end
-    def css
-<<HERE
-table {
-  border-collapse: collapse;
-}
-th {
-  text-align: left;
-  padding-right: 1em;
-  border-bottom: 3px solid #ccc;
-}
-th.active img {
-  display: inline;
-}
-tr.even, tr.odd {
-  background-color: #eee;
-  border-bottom: 1px solid #ccc;
-}
-tr.even, tr.odd {
-  padding: 0.1em 0.6em;
-}
-td.active {
-  background-color: #ddd;
-}
-table td {
-border:1px solid #aaa;
-}
-table tr.line td{
-border-top: 2px solid black;
-}
-
-HERE
-    end
-    
-    def create_uls(level)
-        if @c_level!=level
-            if level>@c_level
-                "<ul>\n" * (level-@c_level)
-            else
-                "</ul>\n" * (@c_level-level)
-            end
-        else
-            ""
-        end
-    end
-    
-    def parse
-        html="<html><head><title>#{@name}</title><style>#{css()}</style></head><body><h1>Report: #{@name}</h1>"
-        if @anchors.size>0
-            html << "<div class='index'>Index</div><ul>"
-            @c_level=1
-            @anchors.each{|name,level,uniq|
-                html << create_uls(level)
-                @c_level=level
-                html << "<li><a href='#"+uniq.to_s+"'>#{name}</a></li>"
-            }
-            html << create_uls(1)
-            html << "</ul></div>"
-        end
-        html+="<div class='section'>"+@partials.join("</div><div class='section'>")+"</div>"
-        html+="</body></html>"
-        html
-    end
-    def save(filename)
-        File.open(filename,"w") {|fp|
-            fp.write(parse)
-        }
-    end
-end
-end