statsample-timeseries 0.0.3 → 0.3.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 2aa5a74001152045a97547f2341b4a03f23ed408
+  data.tar.gz: 9f1c370a27c0b3ffb99d13f0d96eb6bfc6ca93bc
+SHA512:
+  metadata.gz: 75272189f0b3f8f242820dc5693d335abb8d160d9322c36f7e82372d282ad2f7368ddb5fbedea0ebb2afd6bab21559651f8faaf224a02d462758131cd48098e5
+  data.tar.gz: ca16839c93757d521eb4322640b425fb0672187fa4c05d8adc6ed741fcc74271e5d67d074898a757a1814ff2c45d4a3e86e32d4a540d5cb16fc98b7504f7c8d7

data/.gitignore

@@ -0,0 +1,19 @@
+# rcov generated
+coverage
+coverage.data
+
+# rdoc generated
+rdoc
+
+# yard generated
+doc
+.yardoc
+
+# bundler
+.bundle
+
+# jeweler generated
+pkg
+
+Gemfile.lock
+*.gem

data/.travis.yml

@@ -1,13 +1,16 @@
 language: ruby
+
 rvm:
-  - 1.9.2
-  - 1.9.3
   - 2.0.0
-  - jruby-19mode # JRuby in 1.9 mode
-  - rbx-19mode
-#  - 1.8.7
-#  - jruby-18mode # JRuby in 1.8 mode
-#  - rbx-18mode
-
-# uncomment this line if your project needs to run something other than `rake`:
-# script: bundle exec rspec spec
+  - 2.1.0
+  - 2.2.1
+
+script: "bundle exec rake test"
+
+install:
+  - gem install bundler
+  - bundle install
+
+before_install:
+  - sudo apt-get update -qq
+  - sudo apt-get install -y libgsl0-dev

data/Gemfile

@@ -1,21 +1,2 @@
-source "http://rubygems.org"
-
-gem 'statsample', '=1.2.0'
-
-# Add dependencies required to use your gem here.
-# Example:
-   gem "activesupport", "= 3.2.10"
-
-# Add dependencies to develop your gem here.
-# Include everything needed to run rake, tests, features, etc.
-group :development do
-  gem "shoulda", ">= 0"
-  gem "rdoc", "~> 3.12"
-  gem "minitest", "~> 4.7.5"
-  gem "cucumber", ">= 0"
-  gem "bundler", "~> 1.3.5"
-  gem "jeweler", "~> 1.8.4"
-  gem "bio", ">= 1.4.2"
-  gem "rdoc", "~> 3.12"
-  gem 'mocha', '~> 0.14.0'
-end
+source "https://rubygems.org"
+gemspec

data/History.md

@@ -0,0 +1,4 @@
+# 0.1.0
+
+* Statsample-timeseries now dependents on the latest statsample (2) and uses daru for data containers.
+* Removed redundant dependency on cucumber and migrated relevant cucumber tests to Test::Unit.

data/LICENSE.txt

@@ -9,7 +9,7 @@ You *must* read the Contributor Agreement before contributing code to the SciRub
 
 -----
 
-Copyright (c) 2010 - 2013, Ruby Science Foundation
+Copyright (c) 2013, Ankur Goel and Ruby Science Foundation
 All rights reserved.
 
 Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

data/README.md

@@ -0,0 +1,62 @@
+# statsample-timeseries
+
+{<img
+src="https://secure.travis-ci.org/SciRuby/statsample-timeseries.png"
+/>}[http://travis-ci.org/#!/SciRuby/statsample-timeseries]
+
+Statsample-Timeseries is an extension to [Statsample](https://github.com/sciruby/statsample), a suite for advanced statistics with Ruby.
+
+## Description
+
+Statsample-Timeseries is extension of Statsample, and incorporates helpful time series functions, estimation techniques, and modules, such as:
+
+* ACF
+* PACF
+* ARIMA
+* Kalman Filter
+* Log Likelihood
+* Autocovariances
+* Moving Averages
+
+Statsample-Timeseries is part of the [SciRuby project](http://sciruby.com).
+
+## Dependency
+
+Please install [rb-gsl]() which is a Ruby wrapper over GNU Scientific Library.
+It enables us to use various minimization techniques during estimations.
+
+## Installation
+
+`gem install statsample-timeseries`
+
+## Usage
+
+To use the library:
+
+`require 'statsample-timeseries'`
+
+See [Ankur's blog posts](http://ankurgoel.com) for explanations and examples.
+
+## Documentation
+
+The API doc is [online](http://rubygems.org/gems/statsample-timeseries). For more code examples see also the test files in the source tree.
+
+## Contributing
+
+* Fork the project.
+* Create your feature branch
+* Add/Modify code.
+* Write equivalent documentation and *tests*.
+* Run `rake test` to verify that all tests pass.
+* Push your branch.
+* Pull request. :)
+
+## Project home page
+
+For information on the source tree, documentation, issues and how to contribute, see [http://github.com/SciRuby/statsample-timeseries](git@github.com:SciRuby/statsample-timeseries.git).
+
+== Copyright
+
+Copyright (c) 2013 Ankur Goel and the Ruby Science Foundation. See LICENSE.txt for further details.
+
+Statsample is (c) 2009-2013 Claudio Bustos and the Ruby Science Foundation.

data/Rakefile

@@ -1,6 +1,8 @@
 # encoding: utf-8
+require 'rake'
+require 'bundler/gem_tasks'
 
-require 'rubygems'
+# Setup the necessary gems, specified in the gemspec.
 require 'bundler'
 begin
   Bundler.setup(:default, :development)
@@ -9,20 +11,16 @@ rescue Bundler::BundlerError => e
   $stderr.puts "Run `bundle install` to install missing gems"
   exit e.status_code
 end
-require 'rake'
 
-require 'jeweler'
-Jeweler::Tasks.new do |gem|
-  # gem is a Gem::Specification... see http://docs.rubygems.org/read/chapter/20 for more options
-  gem.name = "statsample-timeseries"
-  gem.homepage = "http://github.com/AnkurGel/statsample-timeseries"
-  gem.summary = %Q{TimeSeries modules for Statsample}
-  gem.description = %Q{Statsample-timeseries is an extension to Statsample. It incorporates helpful timeseries functions and modules like ARMA, ARIMA, acf, pacf, lags etc.}
-  gem.email = "ankurgel@gmail.com"
-  gem.authors = ["Ankur Goel", "Claudio Bustos"]
-  # dependencies defined in Gemfile
+desc "Open IRB with statsample-timeseries loaded."
+task :console do
+  require 'irb'
+  require 'irb/completion'
+  $:.unshift File.expand_path("../lib", __FILE__)
+  require 'statsample-timeseries'
+  ARGV.clear
+  IRB.start
 end
-Jeweler::RubygemsDotOrgTasks.new
 
 require 'rake/testtask'
 Rake::TestTask.new(:test) do |test|
@@ -31,14 +29,11 @@ Rake::TestTask.new(:test) do |test|
   test.verbose = true
 end
 
-require 'cucumber/rake/task'
-Cucumber::Rake::Task.new(:features)
-
 task :default => :test
 
 require 'rdoc/task'
 Rake::RDocTask.new do |rdoc|
-  version = File.exist?('VERSION') ? File.read('VERSION') : ""
+  version = Statsample::TimeSeries::VERSION
 
   rdoc.rdoc_dir = 'rdoc'
   rdoc.title = "statsample-timeseries #{version}"

data/lib/statsample-timeseries.rb

@@ -1,18 +1,8 @@
-# Please require your code below, respecting the naming conventions in the
-# bioruby directory tree.
-#
-# For example, say you have a plugin named bio-plugin, the only uncommented
-# line in this file would be 
-#
-#   require 'bio/bio-plugin/plugin'
-#
-# In this file only require other files. Avoid other source code.
-
 require 'statsample'
-require_relative 'statsample-timeseries/timeseries.rb'
+require 'statsample-timeseries/version'
+
+require_relative 'statsample-timeseries/daru_monkeys.rb'
 require_relative 'statsample-timeseries/arima.rb'
 require_relative 'statsample-timeseries/arima/kalman'
 require_relative 'statsample-timeseries/arima/likelihood'
 require_relative 'statsample-timeseries/utility.rb'
-
-

data/lib/statsample-timeseries/arima.rb

@@ -1,4 +1,3 @@
-#require 'debugger'
 require 'statsample-timeseries/arima/kalman'
 require 'statsample-timeseries/arima/likelihood'
 module Statsample
@@ -10,31 +9,31 @@ module Statsample
       Statsample::TimeSeries::ARIMA.new
     end
 
-    class ARIMA < Statsample::Vector
-      include Statsample::TimeSeries
+    class ARIMA
 
       #= Kalman filter on ARIMA model
-      #== Params
+      # == Arguments
       #
       #* *ts*: timeseries object
       #* *p*: AR order
       #* *i*: Integerated part order
       #* *q*: MA order
       #
-      #== Usage
-      # ts = (1..100).map { rand }.to_ts
-      # k_obj = Statsample::TimeSeries::ARIMA.ks(ts, 2, 1, 1)
-      # k_obj.ar
-      # #=> AR's phi coefficients
-      # k_obj.ma 
-      # #=> MA's theta coefficients
+      # == Usage
+      #
+      #   ts = (1..100).map { rand }.to_ts
+      #   k_obj = Statsample::TimeSeries::ARIMA.ks(ts, 2, 1, 1)
+      #   k_obj.ar
+      #   #=> AR's phi coefficients
+      #   k_obj.ma 
+      #   #=> MA's theta coefficients
       #
       #== Returns
-      #Kalman filter object
+      # Kalman filter object
       def self.ks(ts, p, i, q)
         #prototype
         if i > 0
-          ts = ts.diff(i).reject { |x| x.nil? }.to_ts
+          ts = ts.diff(i).reject { |x| x.nil? }
         end
         if Statsample.has_gsl?
           filter = Arima::KalmanFilter.new(ts, p, i, q)
@@ -49,14 +48,15 @@ module Statsample
         #http://en.wikipedia.org/wiki/Autoregressive_model#Definition
         #For finding parameters(to fit), we will use either Yule-walker
         #or Burg's algorithm(more efficient)
+        raise NotImplementedError
       end
 
-      #Converts a linear array into a Statsample vector
+      # Converts a linear array into a Daru vector
       #== Parameters
       #
-      #* *arr*: Array which has to be converted in Statsample vector
+      #* *arr*: Array which has to be converted to Daru vector
       def create_vector(arr)
-        Statsample::Vector.new(arr, :scale)
+        Daru::Vector.new(arr)
       end
 
       #=Yule Walker
@@ -70,7 +70,7 @@ module Statsample
       #==Returns
       #phi and sigma vectors
       def yule_walker(ts, n, k)
-        phi, sigma = Pacf::Pacf.yule_walker(ts, k)
+        phi, sigma = Statsample::TimeSeries::Pacf.yule_walker(ts, k)
         return phi, sigma
         #return ar_sim(n, phi, sigma)
       end
@@ -89,22 +89,23 @@ module Statsample
         intermediate = Pacf::Pacf.levinson_durbin(ts, k)
         phi, sigma = intermediate[1], intermediate[0]
         return phi, sigma
-        #return ar_sim(n, phi, sigma)
       end
 
       #=Autoregressive Simulator
       #Simulates an autoregressive AR(p) model with specified number of
       #observations(n), with phi number of values for order p and sigma.
       #
-      #==Analysis:
-      # [http://ankurgoel.com/blog/2013/07/20/ar-ma-arma-acf-pacf-visualizations/](http://ankurgoel.com/blog/2013/07/20/ar-ma-arma-acf-pacf-visualizations/)
+      # == Analysis
       #
-      #==Parameters:
+      # http://ankurgoel.com/blog/2013/07/20/ar-ma-arma-acf-pacf-visualizations/
+      #
+      # == Parameters
       #* *n*: integer, number of observations
       #* *phi* :array of phi values, e.g: [0.35, 0.213] for p = 2
       #* *sigma*: float, sigma value for error generalization
       #
-      #==Usage
+      # == Usage
+      #
       #  ar = ARIMA.new
       #  ar.ar_sim(1500, [0.3, 0.9], 0.12)
       #    # => AR(2) autoregressive series of 1500 values
@@ -127,14 +128,13 @@ module Statsample
         11.upto(n+11) do |i|
           if i <= phi.size
             #dependent on previous accumulation of x
-            backshifts = create_vector(x[0...i].reverse)
+            backshifts = x[0...i].reverse
           else
             #dependent on number of phi size/order
-            backshifts = create_vector(x[(i - phi.size)...i].reverse)
+            backshifts = x[(i - phi.size)...i].reverse
           end
-          parameters = create_vector(phi[0...backshifts.size])
-
-          summation = (backshifts * parameters).inject(:+)
+          parameters = phi[0...backshifts.size]
+          summation = (backshifts.map.with_index { |e,i| e*parameters[i] }).inject(:+)
           x[i] = summation + err_nor.call()
         end
         x - buffer
@@ -144,70 +144,74 @@ module Statsample
       #Simulates a moving average model with specified number of
       #observations(n), with theta values for order k and sigma
       #
-      #==Parameters
+      # == Parameters
       #* *n*: integer, number of observations
       #* *theta*: array of floats, e.g: [0.23, 0.732], must be < 1
       #* *sigma*: float, sigma value for whitenoise error
       #
-      #==Usage
-      #  ar = ARIMA.new
-      #  ar.ma_sim(1500, [0.23, 0.732], 0.27)
+      # == Usage
+      #   ar = ARIMA.new
+      #   ar.ma_sim(1500, [0.23, 0.732], 0.27)
       #
-      #==Returns
-      #Array of generated MA(q) model
+      # == Returns
+      # Array of generated MA(q) model
+      #
+      # == Notes
+      # n is number of observations (eg: 1000). theta are the model parameters 
+      # containting q values. q is the order of MA
       def ma_sim(n, theta, sigma)
-        #n is number of observations (eg: 1000)
-        #theta are the model parameters containting q values
-        #q is the order of MA
-        mean = theta.to_ts.mean()
+        q = theta.size
+        mean = theta.inject(:+) / q
         whitenoise_gen = Distribution::Normal.rng(0, sigma)
         x = Array.new(n, 0)
-        q = theta.size
         noise_arr = (n+1).times.map { whitenoise_gen.call() }
 
         1.upto(n) do |i|
           #take care that noise vector doesn't try to index -ve value:
           if i <= q
-            noises = create_vector(noise_arr[0..i].reverse)
+            noises = noise_arr[0..i].reverse
           else
-            noises = create_vector(noise_arr[(i-q)..i].reverse)
+            noises = noise_arr[(i-q)..i].reverse
           end
-          weights = create_vector([1] + theta[0...noises.size - 1])
+          weights = [1] + theta[0...noises.size - 1]
 
-          summation = (weights * noises).inject(:+)
+          summation = weights.map.with_index { |e,i| e*noises[i] }.inject(:+)
           x[i] = mean + summation
         end
+
         x
       end
 
       #=ARMA(Autoregressive and Moving Average) Simulator
-      #ARMA is represented by:
-      #http://upload.wikimedia.org/math/2/e/d/2ed0485927b4370ae288f1bc1fe2fc8b.png
-      #This simulates the ARMA model against p, q and sigma.
-      #If p = 0, then model is pure MA(q),
-      #If q = 0, then model is pure AR(p),
-      #otherwise, model is ARMA(p, q) represented by above.
+      # ARMA is represented by:
+      # http://upload.wikimedia.org/math/2/e/d/2ed0485927b4370ae288f1bc1fe2fc8b.png
+      # This simulates the ARMA model against p, q and sigma.
+      # If p = 0, then model is pure MA(q),
+      # If q = 0, then model is pure AR(p),
+      # otherwise, model is ARMA(p, q) represented by above.
       #
-      #==Detailed analysis:
-      # [http://ankurgoel.com/blog/2013/07/20/ar-ma-arma-acf-pacf-visualizations/](http://ankurgoel.com/blog/2013/07/20/ar-ma-arma-acf-pacf-visualizations/)
+      # == Detailed analysis:
+      #
+      # http://ankurgoel.com/blog/2013/07/20/ar-ma-arma-acf-pacf-visualizations/
+      #
+      # == Parameters
       #
-      #==Parameters
       #* *n*: integer, number of observations
       #* *p*: array, contains p number of phi values for AR(p) process
       #* *q*: array, contains q number of theta values for MA(q) process
       #* *sigma*: float, sigma value for whitenoise error generation
       #
-      #==Usage
-      #  ar = ARIMA.new
-      #  ar.arma_sim(1500, [0.3, 0.272], [0.8, 0.317], 0.92)
+      # == Usage
       #
-      #==Returns
-      #array of generated ARMA model values
+      #   ar = ARIMA.new
+      #   ar.arma_sim(1500, [0.3, 0.272], [0.8, 0.317], 0.92)
+      #
+      # == Returns
+      #
+      #  array of generated ARMA model values
       def arma_sim(n, p, q, sigma)
         #represented by :
         #http://upload.wikimedia.org/math/2/e/d/2ed0485927b4370ae288f1bc1fe2fc8b.png
-
-
         whitenoise_gen = Distribution::Normal.rng(0, sigma)
         noise_arr = (n+11).times.map { whitenoise_gen.call() }
 
@@ -216,35 +220,29 @@ module Statsample
 
         11.upto(n+11) do |i|
           if i <= p.size
-            backshifts = create_vector(x[0...i].reverse)
+            backshifts = x[0...i].reverse
           else
-            backshifts = create_vector(x[(i - p.size)...i].reverse)
+            backshifts = x[(i - p.size)...i].reverse
           end
-          parameters = create_vector(p[0...backshifts.size])
-
-          ar_summation = (backshifts * parameters).inject(:+)
+          parameters = p[0...backshifts.size]
+          ar_summation = backshifts.map.with_index { |e,i| e*parameters[i] }.inject(:+)
 
           if i <= q.size
-            noises = create_vector(noise_arr[0..i].reverse)
+            noises = noise_arr[0..i].reverse
           else
-            noises = create_vector(noise_arr[(i-q.size)..i].reverse)
+            noises = noise_arr[(i-q.size)..i].reverse
           end
-          weights = create_vector([1] + q[0...noises.size - 1])
-
-          ma_summation = (weights * noises).inject(:+)
-
+          weights = [1] + q[0...noises.size - 1]
+          ma_summation = (weights.map.with_index { |e,i| e*noises[i] }).inject(:+)
           x[i] = ar_summation + ma_summation
         end
         x - buffer
       end
 
-      #=Hannan-Rissanen for ARMA fit
+      # Hannan-Rissanen for ARMA fit
       def self.hannan(ts, p, q, k)
-        start_params = create_vector(Array.new(p+q+k, 0))
-        ts_dup = ts.dup
-
+        raise NotImplementedError
       end
     end
-
   end
 end