Journal of Futures Markets最新文献

This paper examines the role of analyst commentary sentiment (AS) in enhancing the forecasting of crude oil price volatility. Specifically, we first construct the AS index based on analyst commentaries and develop a volatility index using 5-min high-frequency crude oil price data. We then apply heterogeneous autoregressive (HAR) models and the state-of-the-art deep-learning models to analyze how analyst sentiment improves the forecasting of crude oil price volatility. The results show that the AS index captures significant information, improving forecasting accuracy of crude oil price volatility over medium-term forecasting horizons, especially when deep-learning models are employed. Additionally, deep-learning models significantly improve the forecasting performance during periods of high volatility and negative analyst commentary sentiment, while traditional HAR models perform poorly during this period. Finally, from the perspective of asset allocation, the AS index helps crude oil futures investors to achieve considerable economic returns.

Traditional heterogeneous autoregressive models of realized volatility (HAR-RV) often fail because of the invalidity of residual randomness assumptions, and limitations arise since their reliance on specific data features for volatility characterization. To address these issues, this study constructs uncertain HAR-RV models based on uncertainty theory. Building on this foundation, this study further introduces uncertain quantiles into the modeling framework, develops uncertain quantile HAR-RV models, and provides parameter estimation along with rigorous mathematical proofs. Finally, this study applies the constructed models to volatility forecasting in China's crude oil futures market. Through randomness tests, out-of-sample evaluations, and robustness tests, the limitations of traditional models that lead to failure are systematically validated, and the superior predictive performance of the proposed models across different quantiles is demonstrated. Furthermore, leveraging the unique perspective of uncertainty theory in handling imprecise data, a new perspective for volatility forecasting that uses uncertainty distributions to characterize the daily realized volatility is provided.

We construct a text-based measure of oil market uncertainty using Wall Street Journal front-page articles from January 1986 to December 2022. This measure, the news implied oil volatility (NOVX), is derived from over 100,000 articles and reflects oil market uncertainty during key historical events. NOVX emerges as a significant predictor of oil returns, both in-sample and out-of-sample, outperforming existing predictors like real-time, interest rate, and macroeconomic variables, as well as other news-based indices. Following Manela and Moreira (2017), we decompose NOVX to identify different rare disaster concerns. Among these concerns, those related to government and natural disasters play particularly significant roles in forecasting oil returns. Additionally, we identify an economic mechanism: increases in news-implied oil volatility reduce oil production and economic activity, while increasing oil inventories and decreasing oil prices.

We examine the predictability of commodity futures returns using image-based price patterns extracted from open-high-low-close (OHLC) charts. Applying convolutional neural networks (CNNs) to US commodity futures data, we extract predictive signals without predefined patterns such as momentum or mean reversion. Empirical results demonstrate that image-based predictions enhance predictive accuracy, particularly over short- and medium-term horizons, with 20-day OHLC images yielding the most robust performance. Compared with traditional financial predictors, CNNs capture nonlinear dependencies while retaining unique explanatory power. Panel regressions confirm that image-based predictions are correlated with established return factors. However, transfer learning—from the US to the Chinese markets—proves ineffective in commodity futures markets, highlighting the necessity of market-specific adaptation.

This study investigates the predictive relationship between media sentiment and daily VIX futures returns through a comprehensive textual analysis of financial news articles, blogs, and online discussions. Employing the Loughran and McDonald lexicon-based methodology, we construct sentiment indices derived from an extensive data set of over 700,000 media posts, with a particular emphasis on overnight sentiment effects. Our empirical results demonstrate that negative sentiment extracted from news articles and online discussions exhibits significant predictive power for subsequent VIX futures returns, whereas blog-based sentiment demonstrates comparatively limited efficacy. Notably, this robust predictive relationship persists even during periods of macroeconomic announcements, suggesting that media sentiment captures information beyond traditional economic indicators. Furthermore, we develop sentiment-based trading strategies that yield exceptional performance metrics, generating annualized risk-adjusted returns of 46.32% for news-derived strategies and 104.58% for online discussion-based approaches—substantially outperforming conventional benchmark strategies.

This study investigates the relationships between Shanghai crude oil futures (International Exchange [INE]) and Brent, West Texas Intermediate (WTI), and Oman crude oil futures from March 26, 2018, to January 16, 2024. We identify three long-term equilibrium relationships among the four oil futures at a daily frequency, while a unique equilibrium exists at a weekly frequency. Notably, only INE reacts to deviations from this unique equilibrium on a weekly basis. In terms of volatility, both daily and weekly, INE engages in mutual risk transfer with Brent, WTI, and Oman. However, INE exhibits lower correlations with the other three futures compared with their high correlations with each other. This aligns with the observation that INE is less volatile under extreme shocks, which leads to an arbitrage strategy based on INE and Oman, yielding an annualized return of 22.89% with a maximum drawdown of 18.85%.

We extend classical option-pricing models by adding a neural network correction that captures the intricate curvature of the implied volatility (IV) surface, even in highly nonlinear regions. Using daily SSE 50 ETF option data, we propose a two-stage hybrid framework that first fits a parametric model and then trains a feedforward neural network to correct residual errors. The correction is updated in a rolling out-of-sample procedure and significantly improves IV predictions across multiple horizons. To assess the trading performance of these predictions, we implement a delta-neutral volatility trading strategy. The hybrid approach outperforms benchmark models in both predictive accuracy and trading performance, delivering higher Sharpe ratios and superior risk-adjusted returns. Our results provide new empirical evidence from the Chinese derivatives market and demonstrate that theory-guided machine learning is especially useful for improving the accuracy and applicability of option pricing models.

Using option order imbalance as a proxy for market makers' inventory pressure, we identify a distinct return reversal pattern between option trading activity and underlying asset returns. Specifically, call (put) order imbalances are contemporaneously positively (negatively) associated with underlying returns, followed by rapid reversals in the subsequent period. This reversal stems primarily from temporary price pressures caused by market makers' delta-hedging activities and remains robust after controlling for multiple factors and across various option classifications. Two empirical scenarios reinforce that option order imbalance reflects market makers' inventory risks rather than informed trading. Additional analyses—including event studies, out-of-sample tests, examinations of dynamic hedging behavior, and panel regressions with expanded SSE-listed exchange traded fund option data—further substantiate these findings. Overall, our results emphasize the critical role of market makers as a noninformational trading channel, significantly shaping the relationship between option trading and underlying asset prices.

This study examines the role of after-hours information in forecasting Chinese commodity futures volatility, exploiting the introduction of a night session that potentially facilitates real-time responses to information originating in overseas markets. We generate timely forecasts on future daytime realized volatility for 10 commodity futures, using heterogeneous autoregressive (HAR) models augmented with and without past nights' realized variance measures. Our results reveal significant predictive power, both in-sample and out-of-sample, associated with the night-time realized volatility across markets. In contrast, the inclusion of daily squared overnight returns as an alternative measure provides limited improvements. Furthermore, we document the empirical merit of separately considering the jump and continuous components in the night-session price variation, with its superior performance being most pronounced over long forecasting horizons. The improved statistical accuracy is also shown to be economically meaningful for a risk-averse investor and remains robust to changes in the identification, estimation, and forecasting procedure.