Delta 2.0 - A program for finding dependencies in tables of data

Research output: Contribution to journalArticle

Abstract

A package written in C for identifying variable dependencies in tables of data is presented. The key ingredient is the δ-test, which establishes dependency structures by exploiting the properties of continuous functions using conditional probabilities formed out of the data. The method estimates most relevant variables, embedding dimensions and noise levels. The program, which is self-contained, also includes optional graphical output.

Details

Authors
  • Hong Pi
  • Carsten Peterson
Organisations
External organisations
  • Lund University
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Other Physics Topics
Original languageEnglish
Pages (from-to)293-306
Number of pages14
JournalComputer Physics Communications
Volume83
Issue number2-3
Publication statusPublished - 1994 Dec
Publication categoryResearch
Peer-reviewedYes