SPL File Correlation.spl

SPL standard and specialized toolkits > com.ibm.streams.financial 2.0.0 > com.ibm.streams.financial.utility > Correlation.spl

Content

Functions

• correlationCoefficient(list<float32>, list<float32>): An SPL function is provided in the Financial Services Toolkit for computing the Coefficient of Correlation between two series of variables.

Functions

float32 correlationCoefficient(list<float32> series1, list<float32> series2)

An SPL function is provided in the Financial Services Toolkit for computing the Coefficient of Correlation between two series of variables.

The source code for the correlationCoefficient function can be found in the following file within your installed environment: $STREAMS_INSTALL/toolkits/com.ibm.streams.financial/com.ibm.streams.financial/utility/Correlation.spl.

Note: The two parameters must be lists that have an equal number of members.

Examples

This example demonstrates how to invoke the Correlation function.

 use com.ibm.streams.financial.utility::*;
 composite CCExample {
   graph
     stream<list<flaot32> series1> Series1Source = FileSource(){
       param
         file: "series1.date";
         format: csv;
         initDelay: 5fl;
     }
     stream<list<float32> series2> Series2Source = FileSource(){
       param
         file: "series2.dat";
         format: csv;
         initDelay: 5fl;
     }
     stream<list<float32> s1, list<float32> s2> JoinedSeries =
     Barrier(Series1Source; Series2Source) {
       output
         JoinedSeries: s1 = series1, s2 = series2;
     }
     stream<float32 cc> CC = Functor(JoinedSeries){
       param
         filter: true;
       output
         CC: cc = correlationCoefficient(s1, s2);
     }
     () as SinkOp1 = FileSink(CC){
       param
         file: "CC.result";
         format: csv;
         flush: 1u;
         writePunctuations: false;
     }
}

Parameters

• series1: A list<float32> value, which contains a series of variables.
• series2: A list<float32> value, which contains a series of variables.

Returns

• float32