Price Dynamics and Arbitrage Opportunities in the Palm Oil Futures Markets: Insights from Malaysia and China Using High-Frequency Data

This study investigates the price dynamics and arbitrage opportunities between Malaysia’s Crude Palm Oil Futures (FCPO) and China’s Palm Oil Futures (P.DCE), using high-frequency 5-minute intraday data from January 2019 to December 2023. Employing an integrated analytical framework that combines linear and nonlinear methods including Pearson correlation, E-G cointegration tests, Hurst exponent analysis, wavelet-based multiscale decomposition, threshold vector error correction modeling (TVECM), and transfer entropy we further enhance robustness by incorporating machine learning techniques, specifically Support Vector Regression (SVR) with a linear kernel. Our results reveal an exceptionally strong long-run co-movement between FCPO and P.DCE (Pearson’s r = 0.9882) and confirm the presence of a stable cointegrating relationship from both linear and non-linear perspectives. However, short-term deviations from equilibrium persist, exhibiting mean-reverting behavior (H < 0.5), which creates statistically detectable, albeit transient, arbitrage windows. Nonlinear analyses indicate asymmetric adjustment dynamics and directional information flow, with FCPO maintaining a leading role in price discovery, though P.DCE’s influence has grown significantly. These findings demonstrate that high-frequency statistical arbitrage strategies are feasible in this cross-market context. By bridging a critical gap in the literature using intraday data and advanced analytics, this study offers valuable insights for traders, risk managers, and policymakers, while contributing to the broader discourse on market efficiency and cross-border linkages in agricultural commodity futures.