Efficiency Check Using Cointegration and Machine Learning Approach: Crude Oil Futures Markets
Abstract This study investigates the Indian crude oil futures market efficiency using cointegration tests with an autoregressive distributed lag (ARDL) model. Additionally, machine learning algorithms, support vector regression, and XGBoost are used with auto-regressive and multi-variate approaches to predict the crude oil futures prices. The experimental results found an efficient market and no significant gains can be made by prediction techniques modeled on the data. Moreover, findings suggest that the auto-regressive model outperforms the multi-variate model which is designed based on different market variables of commodity, exchange rate and equity markets.
Paper
Full text
Efficiency Check Using Cointegration and Machine Learning Approach: Crude Oil Futures Markets
Semantic Scholar · Economics · 2021
Abstract
Abstract This study investigates the Indian crude oil futures market efficiency using cointegration tests with an autoregressive distributed lag (ARDL) model. Additionally, machine learning algorithms, support vector regression, and XGBoost are used with auto-regressive and multi-variate approaches to predict the crude oil futures prices. The experimental results found an efficient market and no significant gains can be made by prediction techniques modeled on the data. Moreover, findings suggest that the auto-regressive model outperforms the multi-variate model which is designed based on different market variables of commodity, exchange rate and equity markets.