test_logit.rb in statsample-1.0.0

- old
+ new
@@ -1,21 +1,65 @@
-require(File.dirname(__FILE__)+'/helpers_tests.rb')
-
+require(File.expand_path(File.dirname(__FILE__)+'/helpers_tests.rb'))
+$:.unshift("/home/cdx/dev/rserve-client/lib")
 class StatsampleLogitTestCase < MiniTest::Unit::TestCase
-  def test_logit_1
-    crime=File.dirname(__FILE__)+'/../data/test_binomial.csv'
-    ds=Statsample::CSV.read(crime)
-    lr=Statsample::Regression::Binomial::Logit.new(ds,'y')
-    assert_in_delta(-38.8669,lr.log_likehood,0.001)
-    assert_in_delta(-5.3658,lr.constant,0.001)
+  context Statsample::Regression::Binomial::Logit do
+    should "return correct values for example" do
+      crime=File.dirname(__FILE__)+'/fixtures/test_binomial.csv'
+      ds=Statsample::CSV.read(crime)
+      lr=Statsample::Regression::Binomial::Logit.new(ds,'y')
+      assert_in_delta(-38.8669,lr.log_likehood,0.001)
+      assert_in_delta(-5.3658,lr.constant,0.001)
+  
+      exp_coeffs={"a"=>0.3270,"b"=>0.8147, "c"=>-0.4031}
+      exp_coeffs.each{|k,v|
+        assert_in_delta(v,lr.coeffs[k],0.001)
+      }
+      exp_errors={'a'=>0.4390,'b'=>0.4270,'c'=>0.3819}
+      exp_errors.each{|k,v|
+        assert_in_delta(v,lr.coeffs_se[k],0.001)
+      }
+      assert_equal(7,lr.iterations)
+      end
+    end
+    begin
+      require 'rserve'
+      require 'statsample/rserve_extension'
+      should "return same similat values to as R gml" do
+       
+        r=Rserve::Connection.new
+        ran=Distribution::Normal.rng
+        samples=100
+        a,b,c=ran.call,ran.call,ran.call
+        logit=lambda {|x| Math.exp(x) / (1+Math.exp(x))}
+        
+        x1=Statsample::Vector.new_scale(samples) {ran.call}
+        x2=Statsample::Vector.new_scale(samples) {ran.call}
+        x3=Statsample::Vector.new_scale(samples) {ran.call}
 
-    exp_coeffs={"a"=>0.3270,"b"=>0.8147, "c"=>-0.4031}
-    exp_coeffs.each{|k,v|
-      assert_in_delta(v,lr.coeffs[k],0.001)
-    }
-    exp_errors={'a'=>0.4390,'b'=>0.4270,'c'=>0.3819}
-    exp_errors.each{|k,v|
-      assert_in_delta(v,lr.coeffs_se[k],0.001)
-    }
-    assert_equal(7,lr.iterations)
-  end
+        y= Statsample::Vector.new_scale(samples) {|i| logit.call(x1[i]*a+x2[i]*b+x3[i]*c+ran.call)}
+        # Generate R object
+        ds={'x1'=>x1,'x2'=>x2,'x3'=>x3,'y'=>y}.to_dataset
+        r.assign('ds',ds)
+        r.eval("mylogit<- glm(ds$y~ds$x1+ds$x2+ds$x3, family=binomial(link='logit'), na.action=na.pass)")
+        
+        r_logit=r.eval('summary(mylogit)')        
+        r_coeffs=r_logit.as_list['coefficients'].to_ruby
+        ruby_logit=Statsample::Regression::Binomial::Logit.new(ds,'y')
+
+        assert_in_delta(r_coeffs[0,0],  ruby_logit.constant,1e-4)
+        assert_in_delta(r_coeffs[0,1],  ruby_logit.constant_se,5e-3)
+        
+        %w{x1 x2 x3}.each_with_index do |f,i|
+          assert_in_delta(r_coeffs[i+1,0], ruby_logit.coeffs[f],1e-4)
+          assert_in_delta(r_coeffs[i+1,1], ruby_logit.coeffs_se[f],5e-3)
+          
+        end
+        
+        r.close
+        
+      end
+      
+    rescue LoadError
+      puts "Require rserve extension"
+    
+  end  
 end