Nonlinear granger causality between natural gas and heating oil prices and selected exchange rates

Abstract

Purpose: The currently observed uncertainty in financial markets related to changes taking place in the modern world requires investors to look for tools that allow for good forecasting of the price of financial instruments. The detection of causal relationships may contribute to improving the quality of forecasts by reducing the variance of the prediction error. The aim of the research is to detect nonlinear Granger causality in both directions between selected financial instruments and to check whether the identified relationships are stable over time. Design/methodology/approach: The study of causal relationships between selected financial instruments was carried out using the nonparametric Diks-Panchenko test. This test identifies all types of relationships: linear and nonlinear. Findings: In the first phase of the study, nonlinear Granger causality was tested using the nonparametric Diks-Panchenko test. Six values of lags and two distance measures were used. It is then shown that the significance of the detected relationships has changed in recent years. For this purpose, two directions of causality and three sub-periods were analyzed. Research limitations/implications: Due to the short-term character of the detected relationships, they should be taken into consideration primarily by market participants, to create effective investment portfolios and risk-hedging strategies. Practical implications: Application in making investment decisions on the capital market. Originality/value: The use of information on causal relationships to improve the quality of forecasts related to the energy and currency markets. Keywords: nonlinear Granger causality, Diks-Panchenko test, heating fuel market, exchange rates. Category of the paper: Research paper.