An Empirical Study on Crude Oil Price Fluctuations, Forecasting and Volatility Analysis

Authors:

Madhan Kumar B, II MBA, General, Department of Management Studies, Vels Institute of Science Technology and Advanced Studies (VISTAS) Pallavaram.

Dr. Jayasree Krishnan, Professor & Director, School of Management Studies & Commerce, Vels Institute of Science Technology and Advanced Studies (VISTAS) Pallavaram. directormba@vistas.ac.in, ORCID ID: 0000-0002-4167-0444

Abstract

The continuous instability in the price of crude oil is constantly redefining investment behaviours, as retail and institutional investors move away from the purchase of tangible assets to commodity investments and energy sector exposure. In this paper, an empirical analysis of the main factors that affect WTI crude oil price variability will be done, while time series forecast models for the same period (2000-2023) will be considered. The conventional approach to price forecasting is often based on single macroeconomic variables; however, this paper considers the effect of price and other factors, such as inflation, interest rates, and USD Index, together with structural crises.

The results show that the strength of the US Dollar Index (USD Index) is the most influential variable (β = −0.578, p < .001), then comes inflation (β = 0.506, p < .001) and interest rates (β = −0.300, p < .001). There is also considerable negative influence from crisis dummies like the Global Financial Crisis and the COVID-19 crisis (−$20.31 and −$7.41 respectively per barrel of oil). The multiple regression model accounts for 67.5% of variance in prices (R² = 0.675). The ARIMA(1,1,0) model performs excellently well both in sample R²=0.954 and predication, MAPE =7.19%.

Keywords: Crude Oil Price Fluctuations, Macroeconomic Determinants, ARIMA Forecasting, Investment Impact, Regression Analysis, Commodity Markets, Behavioural Finance.