A Multivariate Distance Nonlinear Causality Test Based on Partial Distance Correlation: A Machine Learning Application to Energy Futures

Quantitative Finance, 2019, 19 (9): 1531-1542.

27 Pages Posted: 23 Sep 2019

See all articles by Germán G. Creamer

Germán G. Creamer

Stevens Institute of Technology, School of Business; Columbia University - Department of Computer Science

Chihoon Lee

Stevens Institute of Technology

Date Written: May 6, 2019

Abstract

This paper proposes a multivariate distance nonlinear causality test (MDNC) using the partial distance correlation in a time series framework. Partial distance correlation as an extension of the Brownian distance correlation calculates the distance correlation between random vectors X and Y controlling for a random vector Z. Our test can detect nonlinear lagged relationships between time series, and when integrated with machine learning methods it can improve the forecasting power. We apply our method as a feature selection procedure and combine it with the support vector machine and random forests algorithms to study the forecast of the main energy financial time series (oil, coal, and natural gas futures). It shows substantial improvement in forecasting the fuel energy time series in comparison to the classical Granger causality method in time series.

Keywords: financial forecasting, lead-lag relationship, nonlinear correlation, energy finance, support vector machine, Brownian partial distance correlation, random forests

Suggested Citation

Creamer, Germán G. and Lee, Chihoon, A Multivariate Distance Nonlinear Causality Test Based on Partial Distance Correlation: A Machine Learning Application to Energy Futures (May 6, 2019). Quantitative Finance, 2019, 19 (9): 1531-1542. . Available at SSRN: https://ssrn.com/abstract=3453353

Germán G. Creamer (Contact Author)

Stevens Institute of Technology, School of Business ( email )

1 Castle Point on Hudson
Hoboken, NJ 07030
United States
2012168986 (Phone)

HOME PAGE: http://www.creamer-co.com

Columbia University - Department of Computer Science ( email )

New York, NY 10027
United States

Chihoon Lee

Stevens Institute of Technology ( email )

Hoboken, NJ 07030
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
44
Abstract Views
137
PlumX Metrics