statsample 0.6.1 → 0.6.2

data/demo/spss_matrix.rb

@@ -1,3 +0,0 @@
-require File.dirname(__FILE__)+"/../lib/statsample"
-ds=Statsample::PlainText.read(File.dirname(__FILE__)+"/../data/tetmat_test.txt", %w{a b c d e})
-puts Statsample::SPSS.tetrachoric_correlation_matrix(ds)

data/demo/strata_proportion.rb

@@ -1,152 +0,0 @@
-require File.dirname(__FILE__)+"/../lib/statsample"
-require 'statsample/multiset'
-require 'statsample/srs'
-require 'statsample/resample'
-require 'gnuplot'
-
-tests=3000
-sample_size=100
-
-a=[1]*50+[0]*950
-b=[1]*900+[0]*100
-a_size=a.size
-b_size=b.size
-av=a.to_vector(:scale)
-bv=b.to_vector(:scale)
-
-ads={'data'=>a.to_vector(:scale)}.to_dataset
-bds={'data'=>b.to_vector(:scale)}.to_dataset
-
-m=Statsample::Multiset.new(['data'])
-m.add_dataset('a',ads)
-m.add_dataset('b',bds)
-ss=Statsample::StratifiedSample.new(m,{'a'=>a.size,'b'=>b.size})
-
-es=[{'N'=>a_size,'n'=>sample_size/2,'s'=>av.standard_deviation_population}, {'N'=>b_size,'n'=>sample_size/2,'s'=>bv.standard_deviation_population}]
-
-esp=[{'N'=>a_size,'n'=>sample_size/2,'p'=>av.proportion(1.0)}, {'N'=>b_size,'n'=>sample_size/2,'p'=>bv.proportion(1.0)}]
-
-
-
-sd_estimated_wr=Statsample::StratifiedSample.standard_error_ksd_wr(es)
-
-sd_estimated_wor = Statsample::StratifiedSample.standard_error_ksd_wor(es)
-
-sd_estimated_wor_p = Statsample::StratifiedSample.proportion_sd_ksd_wor(esp)
-sd_estimated_wr_p = Statsample::StratifiedSample.proportion_sd_ksd_wr(esp)
-
-pop=(a+b).to_vector(:scale)
-s=pop.standard_deviation_population
-
-
-
-
-puts "-------------"
-
-puts "Estadísticos:"
-puts "Mean:"+pop.mean.to_s
-puts "SD:"+s.to_s
-puts "EE con reemplazo:"+Statsample::SRS.standard_error_ksd_wr(s, sample_size, pop.size).to_s
-puts "EE sin reemplazo:"+Statsample::SRS.standard_error_ksd_wor(s, sample_size,pop.size).to_s
-
-
-
-puts "EE estratified con reemplazo:"+sd_estimated_wr.to_s
-puts "EE estratified sin reemplazo:"+sd_estimated_wor.to_s
-puts "EE estratified con reemplazo(p):"+sd_estimated_wr_p.to_s
-puts "EE estratified sin reemplazo(p):"+sd_estimated_wor_p.to_s
-
-sd_with=[]
-sd_without=[]
-sd_strat_wr=[]
-sd_strat_wor_1=[]
-sd_strat_wor_2=[]
-monte_with=Statsample::Resample.repeat_and_save(tests) {
-    sample= pop.sample_with_replacement(sample_size)
-    sd_with.push(Statsample::SRS.standard_error_esd_wr(sample.sds,sample_size,pop.size))
-    sample.mean
-}
-
-
-monte_without=Statsample::Resample.repeat_and_save(tests) {
-    sample= pop.sample_without_replacement(sample_size)
-    sd_without.push(Statsample::SRS.standard_error_esd_wor(sample.sds,sample_size,pop.size))
-    sample.mean
-}
-
-
-
-stratum_wor=Statsample::Resample.repeat_and_save(tests) {
-    a_sample= {'data'=>av.sample_without_replacement(sample_size/2)}.to_dataset
-    b_sample= {'data'=>bv.sample_without_replacement(sample_size/2)}.to_dataset
-    m=Statsample::Multiset.new(['data'])
-    m.add_dataset('a',a_sample)
-    m.add_dataset('b',b_sample)
-    ss=Statsample::StratifiedSample.new(m,{'a'=>a_size,'b'=>b_size})
-    sd_strat_wor_1.push(ss.standard_error_wor('data'))
-    sd_strat_wor_2.push(ss.proportion_sd_esd_wor('data',1.0))
-    ss.mean('data')    
-}.to_vector(:scale)
-
-stratum_wr=Statsample::Resample.repeat_and_save(tests) {
-    a_sample= {'data'=>av.sample_with_replacement(sample_size/2)}.to_dataset
-    b_sample= {'data'=>bv.sample_with_replacement(sample_size/2)}.to_dataset
-    m=Statsample::Multiset.new(['data'])
-    m.add_dataset('a',a_sample)
-    m.add_dataset('b',b_sample)
-    ss=Statsample::StratifiedSample.new(m,{'a'=>a_size,'b'=>b_size})
-    sd_strat_wr.push(ss.standard_error_wr('data'))
-    ss.mean('data')    
-}.to_vector(:scale)
-
-
-
-v_sd_with=sd_with.to_vector(:scale)
-v_sd_without=sd_without.to_vector(:scale)
-v_sd_strat_wr=sd_strat_wr.to_vector(:scale)
-v_sd_strat_wor_1=sd_strat_wor_1.to_vector(:scale)
-v_sd_strat_wor_2=sd_strat_wor_2.to_vector(:scale)
-
-v_with=monte_with.to_vector(:scale)
-v_without=monte_without.to_vector(:scale)
-puts "=============="
-puts "Con reemplazo"
-puts "Mean:"+v_with.mean.to_s
-puts "Sd:"+v_with.sds.to_s
-puts "Sd (estimated):"+v_sd_with.mean.to_s
-puts "=============="
-puts "Sin reemplazo"
-puts "Mean:"+v_without.mean.to_s
-puts "Sd:"+v_without.sds.to_s
-puts "Sd (estimated):"+v_sd_without.mean.to_s
-puts "=============="
-puts "Estratificado Con reemplazo"
-puts "Mean:"+stratum_wr.mean.to_s
-puts "Sd:"+stratum_wr.sds.to_s
-puts "Sd (estimated):"+v_sd_strat_wr.mean.to_s
-
-puts "=============="
-puts "Estratificado Sin reemplazo"
-puts "Mean:"+stratum_wor.mean.to_s
-puts "Sd:"+stratum_wor.sds.to_s
-puts "Sd (estimated scale):"+v_sd_strat_wor_1.mean.to_s
-puts "Sd (estimated prop):"+v_sd_strat_wor_2.mean.to_s
-
-=begin
-
-
-
-
-
-x=[]
-y=[]
-y2=[]
-prev=0
-prev_chi=0
-v.frequencies.sort.each{|k,v1|
-	x.push(k)
-	y.push(prev+v1)
-	prev=prev+v1
-}
-GSL::graph(GSL::Vector.alloc(x),  GSL::Vector.alloc(y))
-=end

data/demo/stratum.rb

@@ -1,141 +0,0 @@
-require File.dirname(__FILE__)+"/../lib/statsample"
-require 'statsample/multiset'
-require 'statsample/srs'
-require 'statsample/resample'
-require 'gnuplot'
-
-tests=3000
-sample_size=50
-
-a=[10]*50+[12]*10+[14]*20+[16]*10+[19]*10
-b=[11000]*50+[11050]*10+[11100]*20+[11300]*10+[11240]*10
-a_size=a.size
-b_size=b.size
-av=a.to_vector(:scale)
-bv=b.to_vector(:scale)
-
-ads={'data'=>a.to_vector(:scale)}.to_dataset
-bds={'data'=>b.to_vector(:scale)}.to_dataset
-
-m=Statsample::Multiset.new(['data'])
-m.add_dataset('a',ads)
-m.add_dataset('b',bds)
-ss=Statsample::StratifiedSample.new(m,{'a'=>a.size,'b'=>b.size})
-
-es=[{'N'=>a_size,'n'=>sample_size/2,'s'=>av.standard_deviation_population}, {'N'=>b_size,'n'=>sample_size/2,'s'=>bv.standard_deviation_population}]
-
-
-
-sd_estimated_wr=Statsample::StratifiedSample.standard_error_ksd_wr(es)
-
-sd_estimated_wor = Statsample::StratifiedSample.standard_error_ksd_wor(es)
-
-
-
-pop=(a+b).to_vector(:scale)
-s=pop.standard_deviation_population
-
-
-
-
-puts "-------------"
-
-puts "Estadísticos:"
-puts "Mean:"+pop.mean.to_s
-puts "SD:"+s.to_s
-puts "EE con reemplazo:"+Statsample::SRS.standard_error_ksd_wr(s, sample_size, pop.size).to_s
-puts "EE sin reemplazo:"+Statsample::SRS.standard_error_ksd_wor(s, sample_size,pop.size).to_s
-
-puts "EE estratified con reemplazo:"+sd_estimated_wr.to_s
-puts "EE estratified sin reemplazo:"+sd_estimated_wor.to_s
-sd_with=[]
-sd_without=[]
-sd_strat_wr=[]
-sd_strat_wor=[]
-monte_with=Statsample::Resample.repeat_and_save(tests) {
-    sample= pop.sample_with_replacement(sample_size)
-    sd_with.push(Statsample::SRS.standard_error_esd_wr(sample.sds,sample_size,pop.size))
-    sample.mean
-}
-
-
-monte_without=Statsample::Resample.repeat_and_save(tests) {
-    sample= pop.sample_without_replacement(sample_size)
-    sd_without.push(Statsample::SRS.standard_error_esd_wor(sample.sds,sample_size,pop.size))
-    sample.mean
-}
-
-
-
-stratum_wor=Statsample::Resample.repeat_and_save(tests) {
-    a_sample= {'data'=>av.sample_without_replacement(sample_size/2)}.to_dataset
-    b_sample= {'data'=>bv.sample_without_replacement(sample_size/2)}.to_dataset
-    m=Statsample::Multiset.new(['data'])
-    m.add_dataset('a',a_sample)
-    m.add_dataset('b',b_sample)
-    ss=Statsample::StratifiedSample.new(m,{'a'=>a_size,'b'=>b_size})
-    sd_strat_wor.push(ss.standard_error_wor('data'))
-    ss.mean('data')    
-}.to_vector(:scale)
-
-stratum_wr=Statsample::Resample.repeat_and_save(tests) {
-    a_sample= {'data'=>av.sample_with_replacement(sample_size/2)}.to_dataset
-    b_sample= {'data'=>bv.sample_with_replacement(sample_size/2)}.to_dataset
-    m=Statsample::Multiset.new(['data'])
-    m.add_dataset('a',a_sample)
-    m.add_dataset('b',b_sample)
-    ss=Statsample::StratifiedSample.new(m,{'a'=>a_size,'b'=>b_size})
-    sd_strat_wr.push(ss.standard_error_wr('data'))
-    ss.mean('data')    
-}.to_vector(:scale)
-
-
-
-v_sd_with=sd_with.to_vector(:scale)
-v_sd_without=sd_without.to_vector(:scale)
-v_sd_strat_wr=sd_strat_wr.to_vector(:scale)
-v_sd_strat_wor=sd_strat_wor.to_vector(:scale)
-
-
-v_with=monte_with.to_vector(:scale)
-v_without=monte_without.to_vector(:scale)
-puts "=============="
-puts "Con reemplazo"
-puts "Mean:"+v_with.mean.to_s
-puts "Sd:"+v_with.sds.to_s
-puts "Sd (estimated):"+v_sd_with.mean.to_s
-puts "=============="
-puts "Sin reemplazo"
-puts "Mean:"+v_without.mean.to_s
-puts "Sd:"+v_without.sds.to_s
-puts "Sd (estimated):"+v_sd_without.mean.to_s
-puts "=============="
-puts "Estratificado Con reemplazo"
-puts "Mean:"+stratum_wr.mean.to_s
-puts "Sd:"+stratum_wr.sds.to_s
-puts "Sd (estimated):"+v_sd_strat_wr.mean.to_s
-
-puts "=============="
-puts "Estratificado Sin reemplazo"
-puts "Mean:"+stratum_wor.mean.to_s
-puts "Sd:"+stratum_wor.sds.to_s
-puts "Sd (estimated):"+v_sd_strat_wor.mean.to_s
-
-p v_without.plot_histogram
-
-=begin
-
-
-
-x=[]
-y=[]
-y2=[]
-prev=0
-prev_chi=0
-v.frequencies.sort.each{|k,v1|
-	x.push(k)
-	y.push(prev+v1)
-	prev=prev+v1
-}
-GSL::graph(GSL::Vector.alloc(x),  GSL::Vector.alloc(y))
-=end

data/demo/t-student.rb

@@ -1,17 +0,0 @@
-require File.dirname(__FILE__)+"/../lib/statsample"
-
-
-tests=3000
-
-r = GSL::Rng.alloc(GSL::Rng::TAUS, 1)
-sample_sizes=[5,10,20,30]
-sample_sizes.each{|sample_size|
-monte=Statsample::Resample.repeat_and_save(tests) {
-    v=[]
-    sample_size.times{|i|
-        v.push(r.ugaussian)
-    }
-    v.to_vector(:scale).mean
-
-}
-}

data/demo/umann.rb

@@ -1,8 +0,0 @@
-require File.dirname(__FILE__)+'/../lib/statsample'
-v1=[1,2,3,4,7,8,9,10,14,15].to_scale
-v2=[5,6,11,12,13,16,17,18,19].to_scale
-u=Statsample::Test::UMannWhitney.new(v1,v2)
-
-puts u.summary
-
-#p Statsample::Test::UMannWhitney.exact_probability_as62(100,100)

data/lib/matrix_extension.rb

@@ -1,92 +0,0 @@
-require 'matrix'
-
-if RUBY_VERSION<="1.9.0"
-  class Vector
-    alias_method :old_coerce, :coerce
-    def coerce(other)
-    case other
-    when Numeric
-      return Matrix::Scalar.new(other), self
-    else
-      raise TypeError, "#{self.class} can't be coerced into #{other.class}"
-    end
-  end
-  end
-end
-class Matrix
-    def rows_sum
-        (0...row_size).collect {|i|
-            row(i).to_a.inject(0) {|a,v| a+v}
-        }
-    end
-    def cols_sum
-        (0...column_size).collect {|i|
-            column(i).to_a.inject(0) {|a,v| a+v}
-        }
-    end
-    def total_sum
-        rows_sum.inject(0){|a,v| a+v}
-    end
-    def row_stochastic
-        rs=rows_sum
-        rows=(0...row_size).collect{|i|
-            (0...column_size).collect {|j|
-                self[i,j].quo(rs[i])
-            }
-        }
-        Matrix.rows(rows,false)
-    end
-    def column_stochastic
-        cs=cols_sum
-        rows=(0...row_size).collect{|i|
-            (0...column_size).collect {|j|
-                self[i,j].quo(cs[j])
-            }
-        }
-        Matrix.rows(rows,false)
-    end
-    def double_stochastic
-        ts=total_sum
-        collect {|i| i.quo(ts)}
-    end
-    # Test if a Matrix is a identity one
-    def identity?
-      if regular?
-      rows=(0...row_size).each{|i|
-      (0...column_size).each {|j|
-        v = self[i,j]
-        return false if (i==j and v!=1) or (i!=j and v!=0)
-      }
-      }
-      true
-      else
-      false
-      end
-    end
-    def to_gsl
-      out=[]
-      self.row_size.times{|i|
-        out[i]=self.row(i).to_a
-      }
-      GSL::Matrix[*out]
-    end
-    def orthogonal?
-      if regular?
-        (self * self.t).identity?
-      else
-        false
-      end
-    end
-end
-
-
-module GSL
-    class Matrix
-        def to_matrix
-            rows=self.size1
-            cols=self.size2
-            out=(0...rows).collect{|i| (0...cols).collect {|j| self[i,j]} }
-            ::Matrix.rows(out)
-        end
-    end
-end