statsample 1.4.3 → 1.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 2a26dc7d632fc818235a92811bee93b9bb1d5349
-  data.tar.gz: 1ab41d3a0ac22838a2805788db87cfaad3bfc8a1
+  metadata.gz: 2bdb8a5e29a75f62af49b094ac026bbc01597707
+  data.tar.gz: 9459ac7bf01e6d20e81a23286dab63f63bdbcea3
 SHA512:
-  metadata.gz: 60c55ef60a46d28f0186d05fc79e7f986a1bc0baa8f6b2cb36decf613ef975409d38d1558cee6288d005d4f5096b4ad066acd9f8729fa164d89316db87f46f42
-  data.tar.gz: c6a4fca050a4a9d881d93a12dd71fa1dcca218414ca17e1af585b42e964aefb6a55acae18f6c6a39194a20b7d5751ae106df7e2935702a42d0147d1b8c101cd3
+  metadata.gz: de03429dbc2c7a1bf0e0acfd0afcebcb55f53630688d635bdfa549eca988bfde40f601e056ae10946f064e86e769bc3be0b50888e1a3209ab243d0253e5b9ed6
+  data.tar.gz: dd71de3f239d8202f3716e4e268f5199cce24e6436e1b017d82970a3bf27da1749bc00a0ab35d030fbd1f63b717aab49c513bb51f8728807a565206cf81d30f1

data/History.txt

@@ -1,3 +1,11 @@
+=== 1.5.0 / 2015-06-11
+  * Made sure all methods work properly with and without GSL.
+  * Statsample works with either rb-gsl or gsl-nmatrix.
+  * Changed the data types of Statsample::Vector from :ordinal, :scale and
+    :nominal to only :numeric and :object. :numeric replaces :ordinal/:scale
+    and :object replaces :nominal. Methods for creating the older data types still
+    exist, but throw a warning prodding the user to use the new methods.
+
 === 1.4.3 / 2015-04-27
   * Removed rb-gsl dependency.
 

data/benchmarks/correlation_matrix_15_variables.rb

@@ -18,7 +18,7 @@ In this test, we test the calculation using #{vars} variables with
 reps 200 #number of repetitions
 
 ds=vars.times.inject({}) {|ac,v|
-ac["x#{v}"]=Statsample::Vector.new_scale(cases) {rand()}
+ac["x#{v}"]=Statsample::Vector.new_numeric(cases) {rand()}
 ac
 }.to_dataset
     

data/benchmarks/correlation_matrix_5_variables.rb

@@ -19,7 +19,7 @@ reps 200 #number of repetitions
 
 
 ds=vars.times.inject({}) {|ac,v|
-ac["x#{v}"]=Statsample::Vector.new_scale(cases) {rand()}
+ac["x#{v}"]=Statsample::Vector.new_numeric(cases) {rand()}
 ac
 }.to_dataset
     

data/benchmarks/correlation_matrix_methods/correlation_matrix.rb

@@ -7,7 +7,7 @@ require 'benchmark'
 def create_dataset(vars,cases) 
   ran=Distribution::Normal.rng
   ds=vars.times.inject({}) {|ac,v|
-    ac["x#{v}"]=Statsample::Vector.new_scale(cases) {ran.call}
+    ac["x#{v}"]=Statsample::Vector.new_numeric(cases) {ran.call}
   ac
   }.to_dataset
 end
@@ -56,7 +56,7 @@ else
 end
 
 
-rs.fields.each {|f| rs[f].type=:scale}
+rs.fields.each {|f| rs[f].type=:numeric}
 
 rs['c_v']=rs.collect {|row| row['cases']*row['vars']}
 

data/examples/dataset.rb

@@ -4,8 +4,8 @@ require 'statsample'
 
 Statsample::Analysis.store(Statsample::Dataset) do
   samples=1000
-  a=Statsample::Vector.new_scale(samples) {r=rand(5); r==4 ? nil: r}
-  b=Statsample::Vector.new_scale(samples) {r=rand(5); r==4 ? nil: r}
+  a=Statsample::Vector.new_numeric(samples) {r=rand(5); r==4 ? nil: r}
+  b=Statsample::Vector.new_numeric(samples) {r=rand(5); r==4 ? nil: r}
 
   ds={'a'=>a,'b'=>b}.to_dataset
   summary(ds)

data/examples/icc.rb

@@ -6,7 +6,7 @@ require 'statsample'
 Statsample::Analysis.store(Statsample::Reliability::ICC) do
 
   size=1000
-  a=Statsample::Vector.new_scale(size) {rand(10)}
+  a=Statsample::Vector.new_numeric(size) {rand(10)}
   b=a.recode{|i|i+rand(4)-2}
   c=a.recode{|i|i+rand(4)-2}
   d=a.recode{|i|i+rand(4)-2}

data/examples/levene.rb

@@ -5,8 +5,8 @@ require 'statsample'
 
 Statsample::Analysis.store(Statsample::Test::Levene) do
 
-  a=[1,2,3,4,5,6,7,8,100,10].to_scale
-  b=[30,40,50,60,70,80,90,100,110,120].to_scale
+  a=[1,2,3,4,5,6,7,8,100,10].to_numeric
+  b=[30,40,50,60,70,80,90,100,110,120].to_numeric
   summary(levene([a,b]))
 end
 

data/examples/parallel_analysis.rb

@@ -15,7 +15,7 @@ f3=rnorm(samples)
 vectors={}
 
 variables.times do |i|
-  vectors["v#{i}"]=samples.times.collect {|nv| f1[nv]*i+(f2[nv]*(15-i))+((f3[nv]*(30-i))*1.5)*rng.call}.to_scale
+  vectors["v#{i}"]=samples.times.collect {|nv| f1[nv]*i+(f2[nv]*(15-i))+((f3[nv]*(30-i))*1.5)*rng.call}.to_numeric
   vectors["v#{i}"].name="Vector #{i}"
 end
 

data/examples/u_test.rb

@@ -4,8 +4,8 @@ require 'statsample'
 
 Statsample::Analysis.store(Statsample::Test::UMannWhitney) do
 
-  a=10.times.map {rand(100)}.to_scale
-  b=20.times.map {(rand(20))**2+50}.to_scale
+  a=10.times.map {rand(100)}.to_numeric
+  b=20.times.map {(rand(20))**2+50}.to_numeric
 
   u=Statsample::Test::UMannWhitney.new(a,b)
   summary u

data/examples/vector.rb

@@ -5,7 +5,7 @@ require 'statsample'
 
 Statsample::Analysis.store(Statsample::Vector) do
 
-  a=Statsample::Vector.new_scale(1000) {r=rand(5); r==4 ? nil: r;}
+  a=Statsample::Vector.new_numeric(1000) {r=rand(5); r==4 ? nil: r;}
   summary a
   b=c(1,2,3,4,6..10)
   summary b

data/examples/velicer_map_test.rb

@@ -21,7 +21,7 @@ Statsample::Analysis.store(Statsample::Factor::MAP) do
   else
     f1[nv]*2 + f2[nv] *3 +rng.call
   end
-  }.to_scale
+  }.to_numeric
   end
   
   

data/lib/statsample.rb

@@ -87,6 +87,32 @@ class Array
       self
     end
   end
+
+  def sum
+    inject(:+)
+  end
+
+  def mean
+    self.sum / size
+  end
+
+  # Calcualte sum of squares
+  def sum_of_squares(m=nil)
+    m ||= mean
+    self.inject(0) {|a,x| a + (x-m).square }
+  end
+
+  # Calculate sample variance
+  def variance_sample(m=nil)
+    m ||= mean
+    sum_of_squares(m).quo(size - 1)
+  end
+
+  # Calculate sample standard deviation
+  def sd
+    m ||= mean
+    Math::sqrt(variance_sample(m))
+  end
 end
 
 def create_test(*args, &_proc)
@@ -141,7 +167,7 @@ module Statsample
       class_variable_get(cv)
     end
   end
-
+  
   create_has_library :gsl
 
   SPLIT_TOKEN = ','
@@ -202,9 +228,9 @@ module Statsample
     # Returns a duplicate of the input vectors, without missing data
     # for any of the vectors.
     #
-    #  a=[1,2,3,6,7,nil,3,5].to_scale
-    #  b=[nil,nil,5,6,4,5,10,2].to_scale
-    #  c=[2,4,6,7,4,5,6,7].to_scale
+    #  a=[1,2,3,6,7,nil,3,5].to_numeric
+    #  b=[nil,nil,5,6,4,5,10,2].to_numeric
+    #  c=[2,4,6,7,4,5,6,7].to_numeric
     #  a2,b2,c2=Statsample.only_valid(a,b,c)
     #  => [#<Statsample::Scale:0xb748c8c8 @data=[3, 6, 7, 3, 5]>,
     #        #<Statsample::Scale:0xb748c814 @data=[5, 6, 4, 10, 2]>,

data/lib/statsample/anova/oneway.rb

@@ -67,9 +67,9 @@ module Statsample
     
     # One Way Anova with vectors
     # Example:
-    #   v1=[2,3,4,5,6].to_scale
-    #   v2=[3,3,4,5,6].to_scale
-    #   v3=[5,3,1,5,6].to_scale
+    #   v1=[2,3,4,5,6].to_numeric
+    #   v2=[3,3,4,5,6].to_numeric
+    #   v3=[5,3,1,5,6].to_numeric
     #   anova=Statsample::Anova::OneWayWithVectors.new([v1,v2,v3])
     #   anova.f
     #   => 0.0243902439024391

data/lib/statsample/anova/twoway.rb

@@ -107,9 +107,9 @@ module Statsample
     
     # Two Way Anova with vectors
     # Example:
-    #   v1=[1,1,2,2].to_scale
-    #   v2=[1,2,1,2].to_scale
-    #   v3=[5,3,1,5].to_scale
+    #   v1=[1,1,2,2].to_numeric
+    #   v2=[1,2,1,2].to_numeric
+    #   v3=[5,3,1,5].to_numeric
     #   anova=Statsample::Anova::TwoWayWithVectors.new(:a=>v1,:b=>v2, :dependent=>v3)
     #
     class TwoWayWithVectors < TwoWay

data/lib/statsample/bivariate.rb

@@ -125,7 +125,7 @@ module Statsample
             nv.push(froms[i]-r*dels[i])
           end
         end
-        nv.to_vector(:scale)
+        nv.to_vector(:numeric)
       end
       # Correlation between v1 and v2, controling the effect of
       # control on both.
@@ -169,7 +169,7 @@ module Statsample
       def covariance_matrix_pairwise(ds)
         cache={}
         matrix=ds.collect_matrix do |row,col|
-          if (ds[row].type!=:scale or ds[col].type!=:scale)
+          if (ds[row].type!=:numeric or ds[col].type!=:numeric)
             nil
           elsif row==col
             ds[row].variance
@@ -215,7 +215,7 @@ module Statsample
         cm=ds.collect_matrix do |row,col|
           if row==col
             1.0
-          elsif (ds[row].type!=:scale or ds[col].type!=:scale)
+          elsif (ds[row].type!=:numeric or ds[col].type!=:numeric)
             nil
           else
             if cache[[col,row]].nil?
@@ -248,7 +248,7 @@ module Statsample
         rows=ds.fields.collect do |row|
           ds.fields.collect do |col|
             v1a,v2a=Statsample.only_valid_clone(ds[row],ds[col])
-            (row==col or ds[row].type!=:scale or ds[col].type!=:scale) ? nil : prop_pearson(t_pearson(ds[row],ds[col]), v1a.size, tails)
+            (row==col or ds[row].type!=:numeric or ds[col].type!=:numeric) ? nil : prop_pearson(t_pearson(ds[row],ds[col]), v1a.size, tails)
           end
         end
         Matrix.rows(rows)
@@ -257,7 +257,7 @@ module Statsample
       # Spearman ranked correlation coefficient (rho) between 2 vectors
       def spearman(v1,v2)
         v1a,v2a=Statsample.only_valid_clone(v1,v2)
-        v1r,v2r=v1a.ranked(:scale),v2a.ranked(:scale)
+        v1r,v2r=v1a.ranked(:numeric),v2a.ranked(:numeric)
         pearson(v1r,v2r)
       end
       # Calculate Point biserial correlation. Equal to Pearson correlation, with
@@ -265,7 +265,7 @@ module Statsample
       def point_biserial(dichotomous,continous)
         ds={'d'=>dichotomous,'c'=>continous}.to_dataset.dup_only_valid
         raise(TypeError, "First vector should be dichotomous") if ds['d'].factors.size!=2
-        raise(TypeError, "Second vector should be continous") if ds['c'].type!=:scale
+        raise(TypeError, "Second vector should be continous") if ds['c'].type!=:numeric
         f0=ds['d'].factors.sort[0]
         m0=ds.filter_field('c') {|c| c['d']==f0}
         m1=ds.filter_field('c') {|c| c['d']!=f0}
@@ -276,7 +276,7 @@ module Statsample
       def tau_a(v1,v2)
         v1a,v2a=Statsample.only_valid_clone(v1,v2)
         n=v1.size
-        v1r,v2r=v1a.ranked(:scale),v2a.ranked(:scale)
+        v1r,v2r=v1a.ranked(:numeric),v2a.ranked(:numeric)
         o1=ordered_pairs(v1r)
         o2=ordered_pairs(v2r)
         delta= o1.size*2-(o2  & o1).size*2

data/lib/statsample/bivariate/pearson.rb

@@ -7,8 +7,8 @@ module Statsample
     # variables.
     # 
     # == Usage
-    #   a = [1,2,3,4,5,6].to_scale
-    #   b = [2,3,4,5,6,7].to_scale
+    #   a = [1,2,3,4,5,6].to_numeric
+    #   b = [2,3,4,5,6,7].to_numeric
     #   pearson = Statsample::Bivariate::Pearson.new(a,b)
     #   puts pearson.r
     #   puts pearson.t

data/lib/statsample/converter/csv.rb

@@ -39,7 +39,7 @@ module Statsample
           end
         end
 
-        convert_to_scale_and_date(ds, fields)
+        convert_to_numeric_and_date(ds, fields)
         ds.update_valid_data
         ds
       end

data/lib/statsample/converters.rb

@@ -18,7 +18,7 @@ module Statsample
         sth.column_info.each {|c|
             vectors[c['name']]=Statsample::Vector.new([])
             vectors[c['name']].name=c['name']
-            vectors[c['name']].type= (c['type_name']=='INTEGER' or c['type_name']=='DOUBLE') ? :scale : :nominal
+            vectors[c['name']].type= (c['type_name']=='INTEGER' or c['type_name']=='DOUBLE') ? :numeric : :object
             fields.push(c['name'])
         }
         ds=Statsample::Dataset.new(vectors,fields)
@@ -106,10 +106,10 @@ module Statsample
           end
         end
       end
-      def convert_to_scale_and_date(ds,fields)
+      def convert_to_numeric_and_date(ds,fields)
         fields.each do |f|
-          if ds[f].can_be_scale?
-            ds[f].type=:scale
+          if ds[f].can_be_numeric?
+            ds[f].type=:numeric
           elsif ds[f].can_be_date?
             ds[f].type=:date
           end
@@ -128,7 +128,7 @@ module Statsample
             next if row==["\x1A"]
             ds.add_case_array(row)
           end
-          convert_to_scale_and_date(ds,fields)
+          convert_to_numeric_and_date(ds,fields)
           ds.update_valid_data
           fields.each {|f|
             ds[f].name=f
@@ -231,7 +231,7 @@ module Statsample
             raise
           end
         end
-        convert_to_scale_and_date(ds, fields)
+        convert_to_numeric_and_date(ds, fields)
         ds.update_valid_data
         fields.each {|f|
           ds[f].name=f
@@ -345,7 +345,7 @@ out
 			# nickname = nickname
 			def variable_definition(carrier,v,name,nickname=nil)
 				nickname = (nickname.nil? ? "" : "nickname=\"#{nickname}\"" )
-				if v.type==:nominal or v.data.find {|d|  d.is_a? String }
+				if v.type==:object or v.data.find {|d|  d.is_a? String }
 					carrier.categorials.push(name)
 					carrier.conversions[name]={}
 					factors=v.factors

data/lib/statsample/dataset.rb

@@ -52,8 +52,8 @@ module Statsample
   #
   # The fast way to create a dataset uses Hash#to_dataset, with
   # field order  as arguments
-  #   v1 = [1,2,3].to_scale
-  #   v2 = [1,2,3].to_scale
+  #   v1 = [1,2,3].to_numeric
+  #   v2 = [1,2,3].to_numeric
   #   ds = {'v1'=>v2, 'v2'=>v2}.to_dataset(%w{v2 v1})
 
   class Dataset
@@ -477,7 +477,7 @@ module Statsample
       }
     end
 
-    def vector_by_calculation(type=:scale)
+    def vector_by_calculation(type=:numeric)
       a=[]
       each do |row|
         a.push(yield(row))
@@ -547,7 +547,7 @@ module Statsample
           a.push(sum.quo(size-invalids))
         end
       end
-      a=a.to_vector(:scale)
+      a=a.to_vector(:numeric)
       a.name=_("Means from %s") % @name
       a
     end
@@ -680,7 +680,7 @@ module Statsample
     end
     # Retrieves a Statsample::Vector, based on the result
     # of calculation performed on each case.
-    def collect(type=:scale)
+    def collect(type=:numeric)
       data=[]
       each {|row|
         data.push yield(row)
@@ -688,7 +688,7 @@ module Statsample
       Statsample::Vector.new(data,type)
     end
     # Same as Statsample::Vector.collect, but giving case index as second parameter on yield.
-    def collect_with_index(type=:scale)
+    def collect_with_index(type=:numeric)
       data=[]
       each_with_index {|row, i|
         data.push(yield(row, i))
@@ -869,7 +869,7 @@ module Statsample
     #   => Vector [4,6]
     def compute(text)
       @fields.each{|f|
-        if @vectors[f].type=:scale
+        if @vectors[f].type=:numeric
           text.gsub!(f,"row['#{f}'].to_f")
         else
           text.gsub!(f,"row['#{f}']")
@@ -958,7 +958,7 @@ module Statsample
       max_n=0
       h=parent_fields.inject({}) {|a,v| a[v]=Statsample::Vector.new([], @vectors[v].type);a }
       # Adding _row_id
-      h['_col_id']=[].to_scale
+      h['_col_id']=[].to_numeric
       ds_vars.push("_col_id")
       @fields.each do |f|
         if f=~re

data/lib/statsample/dominanceanalysis.rb

@@ -7,9 +7,9 @@ module Statsample
   #
   # == Use
   #
-  #  a=1000.times.collect {rand}.to_scale
-  #  b=1000.times.collect {rand}.to_scale
-  #  c=1000.times.collect {rand}.to_scale
+  #  a=1000.times.collect {rand}.to_numeric
+  #  b=1000.times.collect {rand}.to_numeric
+  #  c=1000.times.collect {rand}.to_numeric
   #  ds={'a'=>a,'b'=>b,'c'=>c}.to_dataset
   #  ds['y']=ds.collect{|row| row['a']*5+row['b']*3+row['c']*2+rand()}
   #  da=Statsample::DominanceAnalysis.new(ds,'y')
@@ -279,7 +279,7 @@ module Statsample
     
     def get_averages(averages)
       out={}
-      averages.each{|key,val| out[key]=val.to_vector(:scale).mean }
+      averages.each{|key,val| out[key]=val.to_vector(:numeric).mean }
       out
     end
     # Hash with average for each k size model.

data/lib/statsample/dominanceanalysis/bootstrap.rb

@@ -6,10 +6,10 @@ module Statsample
     # == Usage
     # 
     #  require 'statsample'
-    #  a=100.times.collect {rand}.to_scale
-    #  b=100.times.collect {rand}.to_scale
-    #  c=100.times.collect {rand}.to_scale
-    #  d=100.times.collect {rand}.to_scale
+    #  a=100.times.collect {rand}.to_numeric
+    #  b=100.times.collect {rand}.to_numeric
+    #  c=100.times.collect {rand}.to_numeric
+    #  d=100.times.collect {rand}.to_numeric
     #  ds={'a'=>a,'b'=>b,'c'=>c,'d'=>d}.to_dataset
     #  ds['y']=ds.collect{|row| row['a']*5+row['b']*2+row['c']*2+row['d']*2+10*rand()}
     #  dab=Statsample::DominanceAnalysis::Bootstrap.new(ds2, 'y', :debug=>true)
@@ -182,7 +182,7 @@ module Statsample
           table.row([_("Complete dominance"),"","","","","","",""])
           table.hr
           @pairs.each{|pair|
-            std=@samples_td[pair].to_vector(:scale)
+            std=@samples_td[pair].to_vector(:numeric)
             ttd=da.total_dominance_pairwise(pair[0],pair[1])
             table.row(summary_pairs(pair,std,ttd))
           }
@@ -190,7 +190,7 @@ module Statsample
           table.row([_("Conditional dominance"),"","","","","","",""])
           table.hr
           @pairs.each{|pair|
-            std=@samples_cd[pair].to_vector(:scale)
+            std=@samples_cd[pair].to_vector(:numeric)
             ttd=da.conditional_dominance_pairwise(pair[0],pair[1])
             table.row(summary_pairs(pair,std,ttd))
           
@@ -199,7 +199,7 @@ module Statsample
           table.row([_("General Dominance"),"","","","","","",""])
           table.hr
           @pairs.each{|pair|
-            std=@samples_gd[pair].to_vector(:scale)
+            std=@samples_gd[pair].to_vector(:numeric)
             ttd=da.general_dominance_pairwise(pair[0],pair[1])
             table.row(summary_pairs(pair,std,ttd))
           }
@@ -208,7 +208,7 @@ module Statsample
           table=ReportBuilder::Table.new(:name=>_("General averages"), :header=>[_("var"), _("mean"), _("se"), _("p.5"), _("p.95")])
           
           @fields.each{|f|
-            v=@samples_ga[f].to_vector(:scale)
+            v=@samples_ga[f].to_vector(:numeric)
             row=[@ds[f].name, sprintf("%0.3f",v.mean), sprintf("%0.3f",v.sd), sprintf("%0.3f",v.percentil(5)),sprintf("%0.3f",v.percentil(95))]
             table.row(row)