Programming the statistical procedures from SAS

Understanding the Entropy Metrics in Proc Entropy

Reply
Contributor
Posts: 62

Understanding the Entropy Metrics in Proc Entropy

In reading the documentation for Proc Entropy it's clear that it is still an experimental procedure. This can mean many things but in this case it has consequences for the meaning and interpretation of the metrics as well as the output produced. 

 

For instance, among its Final Information Measures, the procedure produces several metrics for entropy such as Signal Entropy, Noise Entropy, Normed Entropy (Signal), Normed Entropy (Noise), but it's difficult to tell which one is most consistent with Shannon's quite specific definition. Not only that, none of these metrics seem to be available from the ODS tables or, if they are available, they have been unrecognizably renamed.

 

Who in this SAS Community has experience with Proc Entropy and is able to inform as well as correct any misunderstandings or errors in this question and these statements?

 

Thank you.

 

Proc Entropy Output Data

http://support.sas.com/documentation/cdl/en/etsug/63939/HTML/default/viewer.htm#etsug_entropy_sect03...

 

Proc Entropy ODS Table Names

http://support.sas.com/documentation/cdl/en/etsug/63939/HTML/default/viewer.htm#etsug_entropy_sect03...

Ask a Question
Discussion stats
  • 0 replies
  • 45 views
  • 0 likes
  • 1 in conversation