A Mutual Information-Based Network Autoregressive Model for Crude Oil Price Forecasting Using Open-High-Low-Close Prices

The global financial markets are greatly affected by crude oil price movements, indicating the necessity of forecasting their fluctuation and volatility. Crude prices, however, a complex fundamental macroeconomic variable to estimate due nonlinearity, nonstationary, state-of-the-art research in this field demonstrates that conventional methods incapable addressing nonlinear trend changes. Additionally, many parameters involved problem, which adds complexity such prediction. To overcome these obstacles, Mutual Information-Based Network Autoregressive (MINAR) model is developed forecast West Texas Intermediate (WTI) close price. end, open, high, low, (OHLC) prices collected from 1 January 2020 20 July 2022. Afterwards, Information-based distance utilized establish network OHLC prices. MINAR provides basis consider joint effects interactions, autoregressive impact, independent noise establishes an intelligent tool future fluctuations complex, multivariate, noisy environment. measure accuracy performance model, three validation measures, namely, RMSE, MAPE, UMBRAE, applied. results demonstrate proposed outperforms benchmark ARIMA model.