Altered modules hierarchy convention, added more documentation and me…

…thods. Fixes #2
AnkurGel · Aug 31, 2013 · a8e3c0d · a8e3c0d
1 parent 4010b60
commit a8e3c0d
Show file tree

Hide file tree

Showing 3 changed files with 73 additions and 64 deletions.
diff --git a/lib/bio-statsample-timeseries/arima.rb b/lib/bio-statsample-timeseries/arima.rb
@@ -2,6 +2,13 @@
 module Statsample
   # ISSUE: Should be include on module TimeSeries
   module TimeSeries
+
+    def self.arima
+      #not passing (ds,p,i,q) elements for now
+      #will do that once #arima is ready for all modelling
+      Statsample::TimeSeries::ARIMA.new
+    end
+
     class ARIMA < Statsample::Vector
       include Statsample::TimeSeries
       # SUGGESTION: We could use an API similar to R
@@ -209,64 +216,66 @@ def self.hannan(ts, p, q, k)
 
     # ISSUE: This should be included as a module on
     #        Statsample::TimeSeries::Arima::KalmanFilter
-    class KalmanFilter < Statsample::Vector
-      include Statsample::TimeSeries
-
-      #=T
-      #The coefficient matrix for the state vector in state equation
-      # It's dimensions is r+k x r+k
-      #*Parameters*
-      #-_r_::integer, r is max(p, q+1), where p and q are orders of AR and MA respectively
-      #-_k_::integer, number of exogeneous variables in ARMA model
-      #-_q_::integer, The AR coefficient of ARMA model
-
-      #*References*: Statsmodels tsa, Durbin and Koopman Section 4.7
-      def self.T(r, k, p)
-        arr = Matrix.zero(r)
-        params_padded  = Statsample::Vector.new(Array.new(r, 0), :scale)
-
-        params_padded[0...p] = params[k...(p+k)]
-        intermediate_matrix = (r-1).times.map { Array.new(r, 0) }
-        #appending an array filled with padded values in beginning
-        intermediate_matrix[0,0] = [params_padded]
-
-        #now generating column matrix for that:
-        arr = Matrix.columns(intermediate_matrix)
-        arr_00 = arr[0,0]
-
-        #identify matrix substituition in matrix except row[0] and column[0]
-        r.times do |i|
-          arr[r,r] = 1
+    module Arima
+      class KalmanFilter < Statsample::Vector
+        include Statsample::TimeSeries
+
+        #=T
+        #The coefficient matrix for the state vector in state equation
+        # It's dimensions is r+k x r+k
+        #*Parameters*
+        #-_r_::integer, r is max(p, q+1), where p and q are orders of AR and MA respectively
+        #-_k_::integer, number of exogeneous variables in ARMA model
+        #-_q_::integer, The AR coefficient of ARMA model
+
+        #*References*: Statsmodels tsa, Durbin and Koopman Section 4.7
+        def self.T(r, k, p)
+          arr = Matrix.zero(r)
+          params_padded  = Statsample::Vector.new(Array.new(r, 0), :scale)
+
+          params_padded[0...p] = params[k...(p+k)]
+          intermediate_matrix = (r-1).times.map { Array.new(r, 0) }
+          #appending an array filled with padded values in beginning
+          intermediate_matrix[0,0] = [params_padded]
+
+          #now generating column matrix for that:
+          arr = Matrix.columns(intermediate_matrix)
+          arr_00 = arr[0,0]
+
+          #identify matrix substituition in matrix except row[0] and column[0]
+          r.times do |i|
+            arr[r,r] = 1
+          end
+          arr[0,0] = arr_00
+          arr
         end
-        arr[0,0] = arr_00
-        arr
-      end
 
 
-      #=R
-      #The coefficient matrix for the state vector in the observation matrix.
-      #It's dimension is r+k x 1
-      #*Parameters*
-      #-_r_::integer, r is max(p, q+1) where p and q are order of AR and MA respectively
-      #-_k_::integer, number of exogeneous variables in ARMA model
-      #-_q_::integer, The MA order in ARMA model
-      #-_p_::integer, The AR order in ARMA model
-      #*References*: Statsmodels tsa, Durbin and Koopman
-      def self.R(r, k, q, p)
-        arr = Matrix.column_vector(Array.new(r,0.0))
+        #=R
+        #The coefficient matrix for the state vector in the observation matrix.
+        #It's dimension is r+k x 1
+        #*Parameters*
+        #-_r_::integer, r is max(p, q+1) where p and q are order of AR and MA respectively
+        #-_k_::integer, number of exogeneous variables in ARMA model
+        #-_q_::integer, The MA order in ARMA model
+        #-_p_::integer, The AR order in ARMA model
+        #*References*: Statsmodels tsa, Durbin and Koopman
+        def self.R(r, k, q, p)
+          arr = Matrix.column_vector(Array.new(r,0.0))
 
-        #pending - in kind of difficult end here;
-      end
+          #pending - in kind of difficult end here;
+        end
 
-      #=Z
-      #The Z selector matrix
-      #*Parameters*
-      #-_r_::integer, max(p, q+1)
-      #Returns: vector
-      def self.Z(r)
-        arr = Statsample::Vector.new(Array.new(r, 0.0), :scale)
-        arr[0] = 1.0
-        return arr
+        #=Z
+        #The Z selector matrix
+        #*Parameters*
+        #-_r_::integer, max(p, q+1)
+        #Returns: vector
+        def self.Z(r)
+          arr = Statsample::Vector.new(Array.new(r, 0.0), :scale)
+          arr[0] = 1.0
+          return arr
+        end
       end
     end
   end

diff --git a/lib/bio-statsample-timeseries/timeseries.rb b/lib/bio-statsample-timeseries/timeseries.rb
@@ -85,7 +85,8 @@ def pacf(max_lags = nil, method = :yw)
       #Array constituting estimated AR series.
       #
       def ar(n = 1500, k = 1)
-        series = Statsample::ARIMA::ARIMA.new
+        series = Statsample::TimeSeries.arima
+        #series = Statsample::TimeSeries::ARIMA.new
         series.yule_walker(self, n, k)
       end
 

diff --git a/test/test_arima_simulators.rb b/test/test_arima_simulators.rb
@@ -11,10 +11,10 @@ def generate_pacf(simulation)
     ts.pacf
   end
   context("AR(1) simulations") do
-    include Statsample::ARIMA
+    include Statsample
 
     setup do
-      @series = ARIMA.new
+      @series = TimeSeries.arima
       @ar_1_positive = @series.ar_sim(1500, [0.9], 2)
       @ar_1_negative = @series.ar_sim(1500, [-0.9], 2)
 
@@ -73,10 +73,10 @@ def generate_pacf(simulation)
   end
 
   context("AR(p) simulations") do
-    include Statsample::ARIMA
+    include Statsample
 
     setup do
-      @series = ARIMA.new
+      @series = TimeSeries.arima
       @ar_p_positive = @series.ar_sim(1500, [0.3, 0.5], 2)
       @ar_p_negative = @series.ar_sim(1500, [-0.3, -0.5], 2)
     end
@@ -120,9 +120,9 @@ def generate_pacf(simulation)
 
 
   context("MA(1) simulations") do
-    include Statsample::ARIMA
+    include Statsample
     setup do
-      @series = ARIMA.new
+      @series = TimeSeries.arima
       @ma_positive = @series.ar_sim(1500, [0.5], 2)
       @ma_negative = @series.ar_sim(1500, [-0.5], 2)
     end
@@ -153,9 +153,9 @@ def generate_pacf(simulation)
   end
 
   context("MA(q) simulations") do
-    include Statsample::ARIMA
+    include Statsample
     setup do
-      @series = ARIMA.new
+      @series = TimeSeries.arima
       @ma_positive = @series.ar_sim(1500, [0.5, 0.3, 0.2], 2)
       @ma_negative = @series.ar_sim(1500, [-0.5], 2)
     end
@@ -174,8 +174,7 @@ def generate_pacf(simulation)
   end
 
   context("Yule walker estimations") do
-    include Statsample::ARIMA
-    include Statsample::TimeSeries
+    include Statsample
 
     setup do
       @timeseries = 100.times.map { rand }.to_ts