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.

We analyze the negative impact of credit bond issuance ratings on pricing outcomes. Exploiting the exogenous shock from China’s removal of the mandatory bond rating for credit bond issuance, we employ an event study approach to trace out the consequences of lowering issuance thresholds on bond pricing. We show that firms voluntarily retaining bond ratings experienced credit spreads 10.2 percent higher than firms that abandoned ratings. Firms that dropped issuance ratings experienced significant improvements in bond rating quality, reductions in default risk, and enhanced secondary market liquidity. Our findings highlight the critical role that abandoning issuance ratings plays in improving credit bond pricing efficiency under the issuer-paid rating model.

We investigate how internal uncertainty triggered by top executive turnover (TET) disrupts firms' digital transformation. We find that firms with frequent TET are less likely to engage in digital transformation. The underlying mechanism is that top executives adjust long- and short-term risk-taking strategies and alter the direction of earnings management, thereby delaying digital transformation. Top management team characteristics, such as gender, financial expertise, and international experience, along with corporate governance arrangements, significantly shape the relationship between TET and firms' digital transformation. This effect diminishes over time and becomes less pronounced when the analysis is restricted to general manager turnover.

This paper examines whether financial risk resolution infrastructure can reduce banks’ tail risk exposure by strengthening their ability to resolve non-performing loans (NPLs) over the 2013–2025 period. While conventional bank resolution frameworks focus on the distress and restructuring of financial institutions themselves, we study a complementary form of infrastructure that targets the accumulation of impaired assets on bank balance sheets and provides market-based mechanisms for their resolution. Using the staggered approval of local asset management companies (AMCs) across Chinese provinces and a multi-period difference-in-differences design, we find that the establishment of local distressed asset resolution infrastructure significantly reduces banks’ market-perceived tail risk exposure. Local AMCs reduce tail risk by facilitating distressed asset resolution, alleviating credit risk migration pressure, and improving liquidity resilience. Risk resolution involves short-term capital adjustment costs but is also associated with lower exposure to systemic stress and improved operating performance.

Financial turbulence poses substantial challenges to risk management and investment decisionmaking, particularly in emerging markets. This study constructs a novel Chinese Financial Turbulence Index (FTI) using a dictionary-based method augmented by generative artificial intelligence, drawing from a corpus of over 3.6 million financial news articles spanning 2012 to 2023. The FTI exhibits strong responsiveness to macroeconomic conditions and market uncertainty, and significantly predicts negative market returns. To mitigate risks associated with financial turbulence, we develop a hedging framework that integrates scaled principal component analysis (sPCA) with a portfolio-mimicking strategy. The resulting hedging portfolio, which is based on firm-level financial resilience characteristics and complemented by non-equity assets, effectively offsets turbulence-related risks. The FTI and the proposed hedging approach offer timely and practical tools for monitoring and managing financial turbulence.

This paper examines the role of soft power on regional economic outcomes in China over a twelve-year period of rapid income growth. We address the measurement issue of soft power by constructing a dictionary of soft-power-related terms and using textual analysis to process media news. Our findings reveal that soft power has a moderately positive causal effect on regional economic development. We also try to explore the precise channels from the perspectives of the economic composition, inner drive of growth, and growth’s sustainability. Our results are robust to endogeneity concerns, different model specifications, alternative measures of core variables, and additional concerns of external intervention. These results underscore the significance of soft power in regional economic growth and development, emphasizing the need to cultivate soft power resources alongside hard power.

This study investigates how sovereign debt risk affects bank stock performance and how government ESG moderates this relationship. Using monthly data for 578 listed banks in 22 major economies (2008-2022), we find that higher sovereign debt risk reduces returns and increases volatility, with results robust to alternative measures, crisis-period exclusions, and an instrumental variables approach based on sovereign credit rating changes and neighboring countries' sovereign debt risk. We develop a time-varying factor copula framework and a CoEDP-based systemic risk indicator to capture the spillover effects of foreign sovereign debt risk. A China case study shows substantial spillovers into its banking sector, especially after 2017. These findings highlight the need for policymakers to monitor sovereign risks and use government ESG as a mitigation tool.

Based on a data set of earthquakes in China, this study reveals that fund managers exhibit increased pessimism for firms vulnerable to seismic hazards, leading to a marked decrease in net buy volume. Such pessimism is proven to be biased, that is aligned with the salience theory. We further identify the pivotal role of information transmitted through corporate site visits and online interaction in curbing such unwarranted pessimism among fund managers. Information collected through corporate site visits not only alleviates the pessimism of fund managers who conduct site visits by themselves but also benefits fund managers who are not directly involved in site visits. Corporate online interaction serves as the complementary role for site visits in attenuating fund managers' undue pessimism. Our findings contribute to the broader psychology and finance literature on the effect of salience bias on investors' behavior and fill the gap in the literature on how to overcome behavioral bias.

We propose a financial statement (FS) fraud detection framework, called PeerMeta, that makes improvements in all three components of the detection procedure: label measurement, feature set, and detection model. For the label measurement, prior studies mainly adopt FS fraud events that have already been disclosed and confirmed. We construct a new measure based on news coverage that can reflect unrevealed FS fraud behaviors as well. For the feature set, we innovatively add peer factors learned through the business description texts in financial reports. For the detection model, two meta-learning algorithms are applied to aggregate the 19 popular classifiers. The results indicate that the proposed method has amazingly high recall of real fraud cases announced by regulatory authorities, reaching a staggering value of 0.982. We document that all components in PeerMeta contribute to the improvements of FS fraud detection and also showcase the significant economic value of the detection framework and find that recall is more crucial for the economic value than precision.

We study the trading behavior of boundedly rational investors chasing cross-border capital flow in the context of the Shanghai-/Shenzhen-Hong Kong Stock Connect program. The capital flow from Hong Kong to mainland China via this channel, referred to as northbound capital flow (NCF), is widely recognized as smart money in China. We find that mainland Chinese investors exhibit a strong tendency to herd toward stocks with substantial net NCF flows, and investors are especially prone to herd around trade-oriented NCF that pursues short-term gains. We show that the herd behavior is due to enhanced investor sentiment induced by substantial NCF flows. In addition, NCF outflows lead to a more pronounced herding effect, and investment clustering in small-cap stocks is more prominent. Moreover, market panic and the dispersion of stock information mitigate investor herd behavior. Our research provides new insight into the economic consequences of cross-border capital flows in emerging market countries.

Download This Paper Open PDF in Browser Add Paper to My Library Share: Permalink Using these links will ensure access to this page indefinitely Copy URL Copy DOI

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.

Download This Paper Open PDF in Browser Add Paper to My Library Share: Permalink Using these links will ensure access to this page indefinitely Copy URL Copy DOI

We examine seasonality in commodity futures markets using monthly returns for 26 commodities from 1970 through 2023. Only a few commodities in the early years show half-monthly and monthly seasonality. The same-month trading strategy proposed by Keloharju et al. (2016) outperforms the other-month strategy mostly for the subperiod 1990-1999. We then backtest the momentum strategy and momentum with a reversal strategy to compare their performance with that of a seasonality strategy. When these momentum-and-reversal strategies are combined with the seasonality strategy, the result has significant composite returns. Moreover, the seasonal effects in the commodity futures market have weakened in recent years, indicating that the market tends toward efficiency. Nevertheless, our overall results demonstrate that the effects of combination strategies persist and significantly impact the market.