statsample 0.18.0 → 1.0.0

data/lib/statsample/factor/rotation.rb

@@ -27,7 +27,7 @@ module Factor
     attr_accessor :max_iterations
     # Maximum precision    
     attr_accessor :epsilon
-    
+    attr_accessor :use_gsl
     dirty_writer :max_iterations, :epsilon
     dirty_memoize :iterations, :rotated, :component_transformation_matrix, :h2
     
@@ -41,6 +41,7 @@ module Factor
       @epsilon=EPSILON
       @rotated=nil
       @h2=(@matrix.collect {|c| c**2} * Matrix.column_vector([1]*@m)).column(0).to_a
+      @use_gsl=Statsample.has_gsl?
       opts.each{|k,v|
         self.send("#{k}=",v) if self.respond_to? k
       }
@@ -58,11 +59,12 @@ module Factor
     end
     # Start iteration 
     def iterate
-      t=Matrix.identity(@m)
-      b=@matrix.dup
-      h=Matrix.diagonal(*@h2).collect {|c| Math::sqrt(c)}
+      k_matrix=@use_gsl ? GSL::Matrix : ::Matrix
+      t=k_matrix.identity(@m)
+      b=(@use_gsl ? @matrix.to_gsl : @matrix.dup)
+      h=k_matrix.diagonal(*@h2).collect {|c| Math::sqrt(c)}
       h_inverse=h.collect {|c| c!=0 ? 1/c : 0 }
-      bh=h_inverse*b
+      bh=h_inverse * b
       @not_converged=true
       @iterations=0
       while @not_converged
@@ -110,9 +112,14 @@ module Factor
                 t[row_i][i]=tx_rot[row_i]
                 t[row_i][j]=ty_rot[row_i]
               }
-              
-              bh=Matrix.rows(bh)
-              t=Matrix.rows(t)
+              #if @use_gsl
+                bh=k_matrix.[](*bh)
+                t=k_matrix.[](*t)
+              #else
+              #  bh=Matrix.rows(bh)
+              #  t=Matrix.rows(t)
+                
+              #end
             else
               num_pairs=num_pairs-1
               @not_converged=false if num_pairs==0

data/lib/statsample/graph/boxplot.rb

@@ -49,7 +49,6 @@ module Statsample
       # to the anchor location. For example, with the default left alignment, 
       # an angle of Math.PI / 2 causes text to proceed downwards. The default angle is zero.      
       attr_accessor :label_angle
-      
       attr_reader :x_scale, :y_scale
       # Create a new Boxplot.
       # Parameters: Hash of options
@@ -223,11 +222,11 @@ module Statsample
               dot.bottom {|v| y_scale.scale(v)}
               dot.title {|v| v}
             end
-            
-            
-          end          
+          end
         end
+        vis
       end
+      
       # Returns SVG with scatterplot
       def to_svg
         rp=rubyvis_panel

data/lib/statsample/graph/histogram.rb

@@ -120,7 +120,7 @@ module Statsample
         y_scale=Rubyvis::Scale.linear(@minimum_y, @maximum_y).range(0, height - margin_vert)
         
         y_scale.nice
-        max_range=@hist.max
+        
         bins=@hist.bins.times.map {|i|
           {
            :low =>@hist.get_range(i)[0],
@@ -170,6 +170,7 @@ module Statsample
           end
            rubyvis_normal_distribution(pan) if @line_normal_distribution
         end
+        vis
       end
       # Returns SVG with scatterplot
       def to_svg

data/lib/statsample/graph/scatterplot.rb

@@ -193,6 +193,7 @@ module Statsample
             end
           end
         end
+        vis
       end
       # Returns SVG with scatterplot
       def to_svg

data/lib/statsample/matrix.rb

@@ -2,6 +2,9 @@ class ::Vector
   def to_matrix
     ::Matrix.columns([self.to_a])
   end
+  def to_vector
+    self
+  end
 end
 class ::Matrix
   def to_matrix
@@ -28,18 +31,24 @@ class ::Matrix
   if Statsample.has_gsl?
     # Optimize eigenpairs of extendmatrix module using gsl
     def eigenpairs
-      eigval, eigvec= GSL::Eigen.symmv(self.to_gsl)
-      ep=eigval.size.times.map {|i|
-        [eigval[i], eigvec.get_col(i).to_a]
-      }
-      ep.sort{|a,b| a[0]<=>b[0]}.reverse
+      to_gsl.eigenpairs
     end
   end
   
   def eigenvalues
-    eigen[:eigenvalues]
+    eigenpairs.collect {|v| v[0]}
+  end
+  def eigenvectors
+    eigenpairs.collect {|v| v[1]}
+  end
+  def eigenvectors_matrix
+    Matrix.columns(eigenvectors)
   end
+  
+  
 
+  
+  
   def to_gsl
     out=[]
     self.row_size.times{|i|
@@ -55,18 +64,94 @@ module GSL
       def to_matrix
       ::Matrix.columns([self.size.times.map {|i| self[i]}])
       end
+      def to_ary
+        to_a
+      end
+      def to_gsl
+        self
+      end
     end
   end
   class Matrix
     def to_gsl
       self
     end
+    
+    def to_dataset
+      f = (self.respond_to? :fields_y) ? fields_y : column_size.times.map {|i| _("VAR_%d") % (i+1) }
+      ds=Statsample::Dataset.new(f)
+      f.each do |ff|
+        ds[ff].type=:scale
+        ds[ff].name=ff
+      end
+      row_size.times {|i|
+        ds.add_case_array(self.row(i).to_a)
+      }
+      ds.update_valid_data
+      ds.name=self.name if self.respond_to? :name
+      ds
+    end
+    
+    def row_size
+      size1
+    end
+    def column_size
+      size2
+    end
+    def determinant
+      det
+    end
+    def inverse
+      GSL::Linalg::LU.invert(self)
+    end
+    def eigenvalues
+      eigenpairs.collect {|v| v[0]}
+    end
+    def eigenvectors
+      eigenpairs.collect {|v| v[1]}
+    end
+    
+    # Matrix sum of squares
+    def mssq
+      sum=0
+      to_v.each {|i| sum+=i**2}
+      sum
+    end
+    
+    def eigenvectors_matrix
+      eigval, eigvec= GSL::Eigen.symmv(self)
+      GSL::Eigen::symmv_sort(eigval, eigvec, GSL::Eigen::SORT_VAL_DESC)
+      eigvec 
+    end
+    def eigenpairs
+      eigval, eigvec= GSL::Eigen.symmv(self)
+      GSL::Eigen::symmv_sort(eigval, eigvec, GSL::Eigen::SORT_VAL_DESC)
+      @eigenpairs=eigval.size.times.map {|i|
+        [eigval[i],eigvec.get_col(i)]
+      }
+    end
+    
+    #def eigenpairs_ruby
+    #  self.to_matrix.eigenpairs_ruby
+    #end
+    def square?
+      size1==size2
+    end
     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
+    def total_sum
+      sum=0
+      size1.times {|i|
+        size2.times {|j|
+          sum+=self[i,j]
+        }
+      }
+      sum
+    end
   end
 end
 
@@ -122,7 +207,7 @@ module Statsample
     @@covariatematrix=0
 
     # Get type of covariate matrix. Could be :covariance or :correlation
-    def type
+    def _type
       if row_size==column_size
         if row_size.times.find {|i| self[i,i]!=1.0}
           :covariance
@@ -134,11 +219,11 @@ module Statsample
       end
       
     end
-    def type=(t)
+    def _type=(t)
       @type=t
     end
     def correlation
-      if(type==:covariance)
+      if(_type==:covariance)
         matrix=Matrix.rows(row_size.times.collect { |i|
           column_size.times.collect { |j|
             if i==j
@@ -151,7 +236,7 @@ module Statsample
         matrix.extend CovariateMatrix 
         matrix.fields_x=fields_x
         matrix.fields_y=fields_y
-        matrix.type=:correlation
+        matrix._type=:correlation
         matrix
       else
         self
@@ -192,12 +277,17 @@ module Statsample
       columns||=rows
       # Convert all labels on index
       row_index=rows.collect {|v| 
-        v.is_a?(Numeric) ? v : fields_x.index(v)
+        r=v.is_a?(Numeric) ? v : fields_x.index(v)
+        raise "Index #{v} doesn't exists on matrix" if r.nil?
+        r
       }
       column_index=columns.collect {|v| 
-        v.is_a?(Numeric) ? v : fields_y.index(v)
+        r=v.is_a?(Numeric) ? v : fields_y.index(v)
+        raise "Index #{v} doesn't exists on matrix" if r.nil?
+        r
       }
-
+      
+      
       fx=row_index.collect {|v| fields_x[v]}
       fy=column_index.collect {|v| fields_y[v]}
         
@@ -206,14 +296,14 @@ module Statsample
       matrix.extend CovariateMatrix 
       matrix.fields_x=fx
       matrix.fields_y=fy
-      matrix.type=type
+      matrix._type=_type
       matrix
     end
     def report_building(generator)
-      @name||= (type==:correlation ? _("Correlation"):_("Covariance"))+_(" Matrix")
+      @name||= (_type==:correlation ? _("Correlation"):_("Covariance"))+_(" Matrix")
       generator.table(:name=>@name, :header=>[""]+fields_y) do |t|
         row_size.times {|i|
-          t.row([fields_x[i]]+@rows[i].collect {|i1|
+          t.row([fields_x[i]]+row(i).to_a.collect {|i1|
               i1.nil? ? "--" : sprintf("%0.3f",i1).gsub("0.",".")
           })
         }

data/lib/statsample/regression.rb

@@ -22,6 +22,9 @@ module Statsample
     # * Logit Regression:    Statsample::Regression::Binomial::Logit
     # * Probit Regression:    Statsample::Regression::Binomial::Probit
     module Regression
+      
+      LinearDependency=Class.new(Exception)
+      
       # Create a Statsample::Regression::Simple object, for simple regression
       # * x: independent Vector
       # * y: dependent Vector
@@ -78,7 +81,7 @@ module Statsample
         if missing_data==:pairwise
            Statsample::Regression::Multiple::RubyEngine.new(ds,y_var, opts)
         else
-          if Statsample.has_gsl?
+          if Statsample.has_gsl? and false
             Statsample::Regression::Multiple::GslEngine.new(ds, y_var, opts)
           else
             ds2=ds.dup_only_valid

data/lib/statsample/regression/binomial.rb

@@ -49,7 +49,7 @@ module Statsample
         end
         # Constant standard error
         def constant_se
-          i=@fields.index :_constant
+          i=@fields.index "_constant"
           Math::sqrt(@var_cov_matrix[i,i])
         end
         # Regression coefficients

data/lib/statsample/regression/multiple/baseengine.rb

@@ -12,6 +12,8 @@ module Statsample
         attr_reader :valid_cases
         # Number of total cases (dataset.cases)
         attr_reader :total_cases
+        
+        attr_accessor :digits
         def self.univariate?
           true
         end
@@ -23,9 +25,15 @@ module Statsample
           @y_var=y_var
           @r2=nil
           @name=_("Multiple Regression:  %s over %s") % [ ds.fields.join(",") , @y_var]
-          opts.each{|k,v|
+          
+          
+          opts_default={:digits=>3}
+          @opts=opts_default.merge opts
+          
+          @opts.each{|k,v|
             self.send("#{k}=",v) if self.respond_to? k
           }
+          
         end
         # Calculate F Test
         def anova
@@ -159,7 +167,7 @@ module Statsample
           columns.unshift([1.0]*@valid_cases)
           x=Matrix.columns(columns)
           matrix=((x.t*x)).inverse * mse
-          matrix.collect {|i| Math::sqrt(i) if i>0 }
+          matrix.collect {|i| Math::sqrt(i) if i>=0 }
         end
         # T for constant
         def constant_t
@@ -170,24 +178,26 @@ module Statsample
           estimated_variance_covariance_matrix[0,0]
         end
         def report_building(b)
+          di="%0.#{digits}f"
           b.section(:name=>@name) do |g|
             c=coeffs
             g.text _("Engine: %s") % self.class
             g.text(_("Cases(listwise)=%d(%d)") % [@total_cases, @valid_cases])
-            g.text _("R=%0.3f") % r
-            g.text _("R^2=%0.3f") % r2
-            g.text _"R^2 Adj=%0.3f" % r2_adjusted
-            g.text _("Std.Error R=%0.3f") % se_estimate
+            g.text _("R=#{di}") % r
+            g.text _("R^2=#{di}") % r2
+            g.text _"R^2 Adj=#{di}" % r2_adjusted
+            g.text _("Std.Error R=#{di}") % se_estimate
             
-            g.text(_("Equation")+"="+ sprintf('%0.3f',constant) +" + "+ @fields.collect {|k| sprintf('%0.3f%s',c[k],k)}.join(' + ') )
+            g.text(_("Equation")+"="+ sprintf(di,constant) +" + "+ @fields.collect {|k| sprintf("#{di}%s",c[k],k)}.join(' + ') )
             
             g.parse_element(anova)
             sc=standarized_coeffs
+            
             cse=coeffs_se
             g.table(:name=>_("Beta coefficients"), :header=>%w{coeff b beta se t}.collect{|field| _(field)} ) do |t|
-				t.row([_("Constant"), sprintf("%0.3f", constant), "-", constant_se.nil? ? "": sprintf("%0.3f", constant_se), constant_t.nil? ? "" : sprintf("%0.3f", constant_t)])
+				t.row([_("Constant"), sprintf(di, constant), "-", constant_se.nil? ? "": sprintf(di, constant_se), constant_t.nil? ? "" : sprintf(di, constant_t)])
               @fields.each do |f|
-                t.row([f, sprintf("%0.3f", c[f]), sprintf("%0.3f", sc[f]), sprintf("%0.3f", cse[f]), sprintf("%0.3f", c[f].quo(cse[f]))])
+                t.row([f, sprintf(di, c[f]), sprintf(di, sc[f]), sprintf(di, cse[f]), sprintf(di, c[f].quo(cse[f]))])
               end  
             end
           end

data/lib/statsample/regression/multiple/gslengine.rb

@@ -1,131 +1,132 @@
 if Statsample.has_gsl?
-module Statsample
-module Regression
-module Multiple
-# Class for Multiple Regression Analysis
-# Requires rbgsl and uses a listwise aproach.
-# Slower on prediction of values than Alglib, because predict is ruby based.
-# Better memory management on multiple (+1000) series of regression. 
-# If you need pairwise, use RubyEngine
-# Example:
-#
-#   @a=[1,3,2,4,3,5,4,6,5,7].to_vector(:scale)
-#   @b=[3,3,4,4,5,5,6,6,4,4].to_vector(:scale)
-#   @c=[11,22,30,40,50,65,78,79,99,100].to_vector(:scale)
-#   @y=[3,4,5,6,7,8,9,10,20,30].to_vector(:scale)
-#   ds={'a'=>@a,'b'=>@b,'c'=>@c,'y'=>@y}.to_dataset
-#   lr=Statsample::Regression::Multiple::GslEngine.new(ds,'y')
-#            
-class GslEngine < BaseEngine
-  def initialize(ds,y_var, opts=Hash.new)
-    super
-    @ds=ds.dup_only_valid
-    @ds_valid=@ds
-    @valid_cases=@ds_valid.cases
-    @dy=@ds[@y_var]
-    @ds_indep=ds.dup(ds.fields-[y_var])
-    # Create a custom matrix
-    columns=[]
-    @fields=[]
-    max_deps = GSL::Matrix.alloc(@ds.cases, @ds.fields.size)
-    constant_col=@ds.fields.size-1
-    for i in 0...@ds.cases
-        max_deps.set(i,constant_col,1)
-    end
-    j=0
-    @ds.fields.each{|f|
-        if f!=@y_var
-            @ds[f].each_index{|i1|
-                max_deps.set(i1,j,@ds[f][i1])
+  module Statsample
+    module Regression
+      module Multiple
+        # Class for Multiple Regression Analysis
+        # Requires rbgsl and uses a listwise aproach.
+        # Slower on prediction of values than Alglib, because predict is ruby based.
+        # Better memory management on multiple (+1000) series of regression.
+        # If you need pairwise, use RubyEngine
+        # Example:
+        #
+        #   @a=[1,3,2,4,3,5,4,6,5,7].to_vector(:scale)
+        #   @b=[3,3,4,4,5,5,6,6,4,4].to_vector(:scale)
+        #   @c=[11,22,30,40,50,65,78,79,99,100].to_vector(:scale)
+        #   @y=[3,4,5,6,7,8,9,10,20,30].to_vector(:scale)
+        #   ds={'a'=>@a,'b'=>@b,'c'=>@c,'y'=>@y}.to_dataset
+        #   lr=Statsample::Regression::Multiple::GslEngine.new(ds,'y')
+        #
+        class GslEngine < BaseEngine
+          def initialize(ds,y_var, opts=Hash.new)
+            super
+            @ds=ds.dup_only_valid
+            @ds_valid=@ds
+            @valid_cases=@ds_valid.cases
+            @dy=@ds[@y_var]
+            @ds_indep=ds.dup(ds.fields-[y_var])
+            # Create a custom matrix
+            columns=[]
+            @fields=[]
+            max_deps = GSL::Matrix.alloc(@ds.cases, @ds.fields.size)
+            constant_col=@ds.fields.size-1
+            for i in 0...@ds.cases
+              max_deps.set(i,constant_col,1)
+            end
+            j=0
+            @ds.fields.each{|f|
+              if f!=@y_var
+                @ds[f].each_index{|i1|
+                  max_deps.set(i1,j,@ds[f][i1])
+                }
+                columns.push(@ds[f].to_a)
+                @fields.push(f)
+                j+=1
+              end
             }
-            columns.push(@ds[f].to_a)
-            @fields.push(f)
-            j+=1
+            @dep_columns=columns.dup
+            @lr_s=nil
+            c, @cov, @chisq, @status = GSL::MultiFit.linear(max_deps, @dy.gsl)
+            @constant=c[constant_col]
+            @coeffs_a=c.to_a.slice(0...constant_col)
+            @coeffs=assign_names(@coeffs_a)
+            c=nil
+          end
+
+          def _dump(i)
+            Marshal.dump({'ds'=>@ds,'y_var'=>@y_var})
+          end
+          def self._load(data)
+            h=Marshal.load(data)
+            self.new(h['ds'], h['y_var'])
+          end
+
+          def coeffs
+            @coeffs
+          end
+          # Coefficients using a constant
+          # Based on http://www.xycoon.com/ols1.htm
+          def matrix_resolution
+            columns=@dep_columns.dup.map {|xi| xi.map{|i| i.to_f}}
+            columns.unshift([1.0]*@ds.cases)
+            y=Matrix.columns([@dy.data.map  {|i| i.to_f}])
+            x=Matrix.columns(columns)
+            xt=x.t
+            matrix=((xt*x)).inverse*xt
+            matrix*y
+          end
+          def r2
+            r**2
+          end
+          def r
+            Bivariate::pearson(@dy, predicted)
+          end
+          def sst
+            @dy.ss
+          end
+          def constant
+            @constant
+          end
+          def standarized_coeffs
+            l=lr_s
+            l.coeffs
+          end
+          def lr_s
+            if @lr_s.nil?
+              build_standarized
+            end
+            @lr_s
+          end
+          def build_standarized
+            @ds_s=@ds.standarize
+            @lr_s=GslEngine.new(@ds_s,@y_var)
+          end
+          def process_s(v)
+            lr_s.process(v)
+          end
+          # ???? Not equal to SPSS output
+          def standarized_residuals
+            res=residuals
+            red_sd=residuals.sds
+            res.collect {|v|
+              v.quo(red_sd)
+            }.to_vector(:scale)
+          end
+
+          # Standard error for coeffs
+          def coeffs_se
+            out={}
+            evcm=estimated_variance_covariance_matrix
+            @ds_valid.fields.each_with_index do |f,i|
+
+              mi=i+1
+              next if f==@y_var
+              out[f]=evcm[mi,mi]
+            end
+            out
+          end
+
         end
-    }
-    @dep_columns=columns.dup
-    @lr_s=nil
-    c, @cov, @chisq, @status = GSL::MultiFit.linear(max_deps, @dy.gsl)
-    @constant=c[constant_col]
-    @coeffs_a=c.to_a.slice(0...constant_col)
-    @coeffs=assign_names(@coeffs_a)
-    c=nil
-  end
-    
-  def _dump(i)
-    Marshal.dump({'ds'=>@ds,'y_var'=>@y_var})
-  end
-  def self._load(data)
-    h=Marshal.load(data)
-    self.new(h['ds'], h['y_var'])
-  end
-  
-  def coeffs
-    @coeffs
-  end
-  # Coefficients using a constant
-  # Based on http://www.xycoon.com/ols1.htm
-  def matrix_resolution
-    columns=@dep_columns.dup.map {|xi| xi.map{|i| i.to_f}}
-    columns.unshift([1.0]*@ds.cases)
-    y=Matrix.columns([@dy.data.map  {|i| i.to_f}])
-    x=Matrix.columns(columns)
-    xt=x.t
-    matrix=((xt*x)).inverse*xt
-    matrix*y
-  end
-  def r2
-    r**2
-  end
-  def r
-    Bivariate::pearson(@dy, predicted)
-  end  
-  def sst
-    @dy.ss
-  end
-  def constant
-    @constant
-  end
-  def standarized_coeffs
-    l=lr_s
-    l.coeffs
-  end
-  def lr_s
-    if @lr_s.nil?
-        build_standarized
-    end
-    @lr_s
-  end
-  def build_standarized
-    @ds_s=@ds.standarize
-    @lr_s=GslEngine.new(@ds_s,@y_var)
-  end
-  def process_s(v)
-    lr_s.process(v)
-  end
-  # ???? Not equal to SPSS output
-  def standarized_residuals
-    res=residuals
-    red_sd=residuals.sds
-    res.collect {|v|
-        v.quo(red_sd)
-    }.to_vector(:scale)
-  end
-  
-    # Standard error for coeffs
-  def coeffs_se
-    out={}
-    evcm=estimated_variance_covariance_matrix
-    @ds_valid.fields.each_with_index do |f,i|
-      mi=i+1
-      next if f==@y_var
-      out[f]=evcm[mi,mi]
+      end
     end
-    out
-  end
-  
-end
-end
-end
-end # for Statsample
+  end # for Statsample
 end # for if