Rsquared 0.0.1 → 0.0.2

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 217fe13000468eed31354809a3b1cab76081a645
+  data.tar.gz: c631668b39e318e350fb683a1d57bd95227f711e
+SHA512:
+  metadata.gz: 1b74ab721353b68dec614290529e9094b6ddebbd22680b006be69b870ef388bc062a93a3ff76ea1bf3496cc2c441f68aa034125339b96bc80c50918de2e16504
+  data.tar.gz: 53d8faa435ffee52dcac9eb712e37bc02a2695f03b209e9e162a491428713f68dc1677ae621ce059c763f1cb93bad01acc7a4e54ddaa4efd5efe1c0f5e69af8a

data/README.md

@@ -3,8 +3,12 @@
 A full featured Ruby statistics library with assumption verification to make using statistics easy, 
 even with no background.
 
+Travis-CI:
 [![Build Status](https://travis-ci.org/dacohen/Rsquared.png)](https://travis-ci.org/dacohen/Rsquared)
 
+CodeClimate:
+[![Code Climate](https://codeclimate.com/github/dacohen/Rsquared.png)](https://codeclimate.com/github/dacohen/Rsquared)
+
 ## Installation
 
 Add this line to your application's Gemfile:

data/lib/Rsquared/GrubbsTest.rb

@@ -0,0 +1,44 @@
+module Rsquared
+##
+  # Tests for outliers on either side of the data
+  # grubbs = Rsquared::GrubbsTest.new(data)
+  # grubbs.significant? => Boolean
+  #
+
+  class GrubbsTest
+	  ##
+	  # Initializes the Test object with an array of numerical data
+	  #
+
+	  def initialize(data)
+	     @data = data.sort
+   	     @gstat = [((@data.mean - @data.min)/@data.stddev).abs, ((@data.mean - @data.max)/@data.stddev).abs].max
+	  end
+	  
+	  ##
+	  # Returns a boolean indicating the significance of the test at the 5% level
+	  #	  
+
+	  def significant?(alpha=0.05)
+	     if @gstat > Helper::grubbscv(@data.length, alpha) then
+	     	return true
+	     else
+		return false
+	     end
+	  end
+
+	  def inspect
+	      significant?
+	  end
+
+	  ##
+	  # Returns the test statistic as a float
+	  #
+
+	  def statistic
+	      @gstat
+	  end
+
+	  alias_method :outlier?, :significant?	      
+  end
+end

data/lib/Rsquared/KSTest.rb

@@ -0,0 +1,60 @@
+module Rsquared
+##
+  # KSTest implements the Kolomogorov-Smirnov test for normality
+  # kstest = Rsquared::KSTest.new(data)
+  # kstest.normal? => Boolean, indicates normality of data at 5% confidence
+  # 
+
+  class KSTest
+  	##
+	# Intitializes the test object with an array of numerical data
+	#
+
+  	def initialize(data)
+	    @data = data.std.sort!
+	    fn = 0
+	    d = []
+	    range = @data.max - @data.min
+	    @data.each_with_index do |x, i|
+	    	# Calculate Fn
+		fn = i + 1
+		d[i] = fn/@data.length.to_f - Distribution::Normal::cdf(x)
+		fn = 0.0
+	    end
+	    @ksstat = d.max
+	    return @ksstat
+	 end
+	 
+	 ##
+	 # Returns a boolean indiciating the significance of the test a the 5% level
+	 #
+
+	 def significant?
+	     if @ksstat > Helper::kscv(@data.length) then
+	     	return true
+	     else
+		return false
+	     end
+	 end
+
+	 ##
+	 # Returns logical opposite of significance
+	 #
+
+	 def normal?
+	     !self.significant?
+	 end
+
+	 def inspect
+	     significant?
+	 end
+	 
+	 ##
+	 # Returns the test statistic
+	 #
+
+	 def statistic
+	     @ksstat
+	 end
+  end
+end

data/lib/Rsquared/PropTest.rb

@@ -0,0 +1,29 @@
+module Rsquared
+
+       class PropTest < StatTest
+       	     def initialize(data, p0, sided)
+	     	 @data = data
+		 @p0 = p0
+		 @sided = sided
+
+		 if (@data.length*@p0 < 10.0) or (@data.length*(1.0-@p0) < 10.0) then
+		    raise AssumptionError, "The number of successes or failures prediced by the proportion is too small"
+		 end
+		 
+		 @stderr = Math.sqrt((@p0*(1.0-@p0))/@data.length)
+		 @pstat = ((@data.sum.to_f/@data.length.to_f) - @p0)/@stderr
+		 @pvalue = Distribution::Normal::cdf(@pstat)
+		 self.setSidedness!(@sided)
+	    end
+
+
+	    ##
+	    # Returns the z-statistic
+	    
+	    def statistic
+	    	@pstat
+            end
+
+	    ## significant?, inspect implemented by inhertance
+	end
+end

data/lib/Rsquared/StatTest.rb

@@ -0,0 +1,31 @@
+module Rsquared
+       class StatTest
+       	     attr_accessor :pvalue
+       	     def significant?(alpha=0.05)
+	     	 if @pvalue < alpha then
+		    return true
+		 else
+		    return false
+		 end
+	     end
+	     
+	     def inspect
+	     	 @pvalue
+	     end
+
+	     ##
+	     # Modifies p-value to account for tails and/or two-sided tests
+	     #
+	
+	     def setSidedness!(sided)
+	     	 if sided == Upper.tail then
+	       	    @pvalue = 1.0-@pvalue
+	   	 elsif sided == Two.sided then
+	       	    @pvalue = [(1.0-@pvalue)*2.0, @pvalue*2.0].min
+	    	end
+             end
+       end
+
+       class AssumptionError < StandardError
+       end
+end

data/lib/Rsquared/TTest.rb

@@ -0,0 +1,44 @@
+module Rsquared
+##
+  # Tests for deviation of sample mean from expected mean
+  # ttest = Rsquared::TTest.new(data, mu0, sided)
+  # mu0 is the expected value of the sample mean
+  # Supply Rsquared::Upper.tail, Rsquared::Lower.tail or Rsquared::Two.sided
+  # Use Upper.tail when you suspect that the sample mean will be greater than the expected mean
+  # Use Lower.tail when you suspect that the sample mean will be smaller than the expected mean
+  # Use Two.sided when you suspect neither
+
+  class TTest < StatTest
+  	##
+	# Initializes the TTest object with the supplied arguments
+	#
+
+  	def initialize(data, mu0, sided)
+	    @data = data
+  	    @mu0 = mu0
+	    @sided = sided
+	    
+	    if KSTest.new(@data).significant? and @data.length < 40 then
+	       raise AssumptionError, "The data is not close enough to a normal distribution for such a small sample size"
+	    end
+	    if GrubbsTest.new(@data).outlier? then
+	       raise AssumptionError, "Your data has one or more outliers, which the T-Distribution cannot handle"
+	    end
+
+	    @tstat = (@data.mean - @mu0)/(data.stddev/Math.sqrt(@data.length))
+	    @pvalue = Distribution::T::cdf(@tstat, @data.length-1)
+	    self.setSidedness!(@sided)
+	 end
+
+	 ##
+	 # Returns the t-statistic
+	 #
+
+	 def statistic
+	     @tstat
+	 end
+
+	 ## significant?, inspect implemented by inheritance
+
+   end
+end

data/lib/Rsquared/version.rb

@@ -1,3 +1,3 @@
 module Rsquared
-  VERSION = "0.0.1"
+  VERSION = "0.0.2"
 end

data/lib/constants.rb

@@ -18,7 +18,4 @@ module Rsquared
 	     	 return true
 	     end
        end
-
-       class AssumptionError < StandardError
-       end
 end

data/lib/rsquared.rb

@@ -5,169 +5,13 @@ require "constants"
 require "complex"
 require "enumerableext.rb"	      
 
-module Rsquared
-
-  ##
-  # KSTest implements the Kolomogorov-Smirnov test for normality
-  # kstest = Rsquared::KSTest.new(data)
-  # kstest.normal? => Boolean, indicates normality of data at 5% confidence
-  # 
-
-  class KSTest
-  	##
-	# Intitializes the test object with an array of numerical data
-	#
-
-  	def initialize(data)
-	    @data = data.std.sort!
-	    fn = 0
-	    d = []
-	    range = @data.max - @data.min
-	    @data.each_with_index do |x, i|
-	    	# Calculate Fn
-		fn = i + 1
-		d[i] = fn/@data.length.to_f - Distribution::Normal::cdf(x)
-		fn = 0.0
-	    end
-	    @ksstat = d.max
-	    return @ksstat
-	 end
-	 
-	 ##
-	 # Returns a boolean indiciating the significance of the test a the 5% level
-	 #
-
-	 def significant?
-	     if @ksstat > Helper::kscv(@data.length) then
-	     	return true
-	     else
-		return false
-	     end
-	 end
-
-	 ##
-	 # Returns logical opposite of significance
-	 #
-
-	 def normal?
-	     !self.significant?
-	 end
-
-	 def inspect
-	     significant?
-	 end
-	 
-	 ##
-	 # Returns the test statistic
-	 #
-
-	 def statistic
-	     @ksstat
-	 end
-  end
-
-  ##
-  # Tests for outliers on either side of the data
-  # grubbs = Rsquared::GrubbsTest.new(data)
-  # grubbs.significant? => Boolean
-  #
-
-  class GrubbsTest
-	  ##
-	  # Initializes the Test object with an array of numerical data
-	  #
-
-	  def initialize(data)
-	     @data = data.sort
-   	     @gstat = [((@data.mean - @data.min)/@data.stddev).abs, ((@data.mean - @data.max)/@data.stddev).abs].max
-	  end
-	  
-	  ##
-	  # Returns a boolean indicating the significance of the test at the 5% level
-	  #	  
-
-	  def significant?(alpha=0.05)
-	     if @gstat > Helper::grubbscv(@data.length, alpha) then
-	     	return true
-	     else
-		return false
-	     end
-	  end
-
-	  def inspect
-	      significant?
-	  end
-
-	  ##
-	  # Returns the test statistic as a float
-	  #
-
-	  def statistic
-	      @gstat
-	  end
+require 'Rsquared/StatTest'
+require 'Rsquared/KSTest'
+require 'Rsquared/GrubbsTest'
+require 'Rsquared/TTest'
+require 'Rsquared/PropTest'
 
-	  alias_method :outlier?, :significant?	      
-  end
-  
-  ##
-  # Tests for deviation of sample mean from expected mean
-  # ttest = Rsquared::TTest.new(data, mu0, sided)
-  # mu0 is the expected value of the sample mean
-  # Supply Rsquared::Upper.tail, Rsquared::Lower.tail or Rsquared::Two.sided
-  # Use Upper.tail when you suspect that the sample mean will be greater than the expected mean
-  # Use Lower.tail when you suspect that the sample mean will be smaller than the expected mean
-  # Use Two.sided when you suspect neither
-
-  class TTest
-  	##
-	# Initializes the TTest object with the supplied arguments
-	#
-
-  	def initialize(data, mu0, sided)
-	    @data = data
-  	    @mu0 = mu0
-	    @sided = sided
-	    
-	    if KSTest.new(@data).significant? and @data.length < 40 then
-	       raise AssumptionException, "The data is not close enough to a normal distribution for such a small sample size"
-	    end
-	    if GrubbsTest.new(@data).outlier? then
-	       raise AssumptionException, "Your data has one or more outliers, which the T-Distribution cannot handle"
-	    end
-
-	    @tstat = (@data.mean - @mu0)/(data.stddev/Math.sqrt(@data.length))
-	    @pvalue = Distribution::T::cdf(@tstat, @data.length-1)
-	    if @sided == Upper.tail then
-	       @pvalue = 1.0-@pvalue
-	    elsif @sided == Two.sided then
-	       @pvalue = [(1.0-@pvalue)*2.0, @pvalue*2.0].min
-	    end
-	 end
-
-	 def inspect
-	     @pvalue
-	 end
-
-	 ##
-	 # Returns the t-statistic
-	 #
-
-	 def statistic
-	     @tstat
-	 end
-
-	 ##
-	 # Checks for significance at the supplied alpha level
-	 #
-
-	 def significant?(alpha=0.05)
-	     if @pvalue < alpha then
-	     	return true
-	     else
-		return false
-	     end
-	 end
-  end
+module Rsquared
 
   ##
   # The Helper module implements uncommon statistical functions directly
@@ -208,6 +52,19 @@ module Rsquared
 	    return ((n-1)/Math.sqrt(n))*Math.sqrt(tcv**2/((n-2)+tcv**2))
 	end
  
+
+	##
+	# Modifies p-value to account for tails and/or two-sided tests
+	#
+	
+	def adjustForSided(pvalue, sided)
+	    if sided == Upper.tail then
+	       return 1.0-pvalue
+	    elsif sided == Two.sided then
+	       return [(1.0-pvalue)*2.0, pvalue*2.0].min
+	    end
+        end
+
 	 module_function :kscv, :grubbscv
   end
 end

data/test/rsquared_tests.rb

@@ -8,6 +8,7 @@ module Test::Unit::Assertions
 end
 
 $data = [-105, 135, 40, 90, -55, -85, 70, 180, 140, -10, -105, 40, 185, -90, -90, 80, 70, -155, 345, 250, 10, -135, 80, 85, -40, 250, -20, 35, 305, -135]
+$propdata = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0]
 
 class RsquaredTests < Test::Unit::TestCase
 
@@ -27,6 +28,16 @@ class RsquaredTests < Test::Unit::TestCase
 	  assert ttest.significant?
       end
 
+      def test_PropTest
+      	  proptest = Rsquared::PropTest.new($propdata, 0.5, Rsquared::Upper.tail)
+	  assert_in_delta 0.080757, proptest.inspect, 0.001
+	  refute proptest.significant?
+	  
+	  assert_raise(Rsquared::AssumptionError) do
+	  	proptest = Rsquared::PropTest.new([1, 1, 1, 1, 1, 0, 0, 0, 0, 0], 0.5, Rsquared::Upper.tail)
+	  end
+      end
+
       def test_Grubbs
       	  grubbs = Rsquared::GrubbsTest.new($data)
 	  assert_in_delta 2.21, grubbs.statistic, 0.01

metadata

@@ -1,75 +1,64 @@
---- !ruby/object:Gem::Specification 
+--- !ruby/object:Gem::Specification
 name: Rsquared
-version: !ruby/object:Gem::Version 
-  hash: 29
-  prerelease: 
-  segments: 
-  - 0
-  - 0
-  - 1
-  version: 0.0.1
+version: !ruby/object:Gem::Version
+  version: 0.0.2
 platform: ruby
-authors: 
+authors:
 - Daniel Cohen
 autorequire: 
 bindir: bin
 cert_chain: []
-
-date: 2013-06-28 00:00:00 Z
-dependencies: 
-- !ruby/object:Gem::Dependency 
+date: 2013-06-29 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
   name: bundler
-  prerelease: false
-  requirement: &id001 !ruby/object:Gem::Requirement 
-    none: false
-    requirements: 
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
     - - ~>
-      - !ruby/object:Gem::Version 
-        hash: 9
-        segments: 
-        - 1
-        - 3
-        version: "1.3"
+      - !ruby/object:Gem::Version
+        version: '1.3'
   type: :development
-  version_requirements: *id001
-- !ruby/object:Gem::Dependency 
-  name: rake
   prerelease: false
-  requirement: &id002 !ruby/object:Gem::Requirement 
-    none: false
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        hash: 3
-        segments: 
-        - 0
-        version: "0"
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.3'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
   type: :development
-  version_requirements: *id002
-- !ruby/object:Gem::Dependency 
-  name: distribution
   prerelease: false
-  requirement: &id003 !ruby/object:Gem::Requirement 
-    none: false
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        hash: 3
-        segments: 
-        - 0
-        version: "0"
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: distribution
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
   type: :development
-  version_requirements: *id003
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
 description: A full-featured Ruby statistics library with assumption verification
-email: 
+email:
 - dcohen@gatech.edu
 executables: []
-
 extensions: []
-
 extra_rdoc_files: []
-
-files: 
+files:
 - .travis.yml
 - Gemfile
 - LICENSE.txt
@@ -77,48 +66,43 @@ files:
 - Rakefile
 - Rsquared.gemspec
 - install.sh
+- lib/Rsquared/GrubbsTest.rb
+- lib/Rsquared/KSTest.rb
+- lib/Rsquared/PropTest.rb
+- lib/Rsquared/StatTest.rb
+- lib/Rsquared/TTest.rb
 - lib/Rsquared/version.rb
 - lib/constants.rb
 - lib/enumerableext.rb
 - lib/rsquared.rb
-- lib/version.rb
 - test/enumerable_tests.rb
 - test/helper_tests.rb
 - test/rsquared_tests.rb
 homepage: https://github.com/dacohen/Rsquared
-licenses: 
+licenses:
 - MIT
+metadata: {}
 post_install_message: 
 rdoc_options: []
-
-require_paths: 
+require_paths:
 - lib
-required_ruby_version: !ruby/object:Gem::Requirement 
-  none: false
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      hash: 3
-      segments: 
-      - 0
-      version: "0"
-required_rubygems_version: !ruby/object:Gem::Requirement 
-  none: false
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      hash: 3
-      segments: 
-      - 0
-      version: "0"
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
 requirements: []
-
 rubyforge_project: 
-rubygems_version: 1.8.24
+rubygems_version: 2.0.3
 signing_key: 
-specification_version: 3
+specification_version: 4
 summary: Provides statistical distributions, tests and verifies relevant assumptions
-test_files: 
+test_files:
 - test/enumerable_tests.rb
 - test/helper_tests.rb
 - test/rsquared_tests.rb