Fuwei Jiang’s research while affiliated with Xiamen University and other places

... After 2001, the effect of sentiment-driven investors and their stock misvaluation in the USA are marginal. According to Jiang et al. [27], mispricing is a relevant factor in capturing stock returns in 33 markets. ...

Reference:

Stock Mispricing and Firm Innovation: Evidence from an Emerging Equity Market

... Additionally, we are the first to validate that deep hedging empirically performs equally well for put options as compared to call options. This paper is related to the extensive application of machine learning in option pricing and hedging (Amilon 2003;Garcia and Gençay 2000;Hutchinson et al. 1994;Ivașcu 2021;Huang et al. 2022, Hong et al. 2024. Instead of deriving option price from no-arbitrage or other economic and statistical assumptions, these studies consider option price as a functional mapping output based on information about the option and the underlying asset (Ivașcu 2021). ...

Reference:

Black‐Scholes Meet Imitation Learning: Evidence From Deep Hedging in China

... Empirical work further consolidates the role of fuzzy logic and hybrid modeling, presenting evidence-based improvements in forecasting accuracy and decision support capabilities. Jiang et al. [22] demonstrated the superior performance of machine learning algorithms, particularly deep learning, in predicting stock price crash risks by leveraging firm-specific characteristics, with a focus on profitability and value versus growth features, offering nuanced economic interpretations and highlighting significant applications within state-owned enterprises and low economic policy uncertainty periods in the Chinese stock market. Chen et al. [21] proposed a fuzzy time-series model incorporating the Fibonacci sequence to improve stock price forecasting accuracy. ...

Reference:

Stock price forecasting through symbolic dynamics and state transition graphs with a convolutional recurrent neural network architecture

... The most critical reason behind this choice is the existence of the momentum feature in the 4-FM. Integrating the momentum factor gives the 4-FM a more holistic view of what may drive the stock returns (Ma et al., 2024). It also improves the model's capacity to explain performance in short-term investment (Fama and French, 2015), which is often important to investors and portfolio managers. ...

Reference:

Does the Carhart (1997) Four-Factor Model Still Applicable? Evidence from Stocks Listed on Bursa Malaysia

... While much of the existing literature emphasizes how shocks stemming from U.s. economic indicators (Jiang et al., 2023;shi & Wang, 2023shi & Wang, ), trade relations (sheikh et al., 2024 and climate policy uncertainties (kayani et al., 2024) influence equity markets, fewer studies investigate other dimensions of shock propagation. For example, Yunus (2023) examines how gDP-related shocks within the U.s. economy are transmitted to financial markets in emerging countries. ...

Reference:

Shock transmission from global financial stress, bitcoin sentiment indices, U.S. and euro financial market uncertainty toward the GCC stock volatility

... Similarly, Nyakurukwa and Seetharam (2023) analyzed textual sentiment from news and Twitter to assess price deviations in dual-listed stocks, while Costola et al. (2023) used machine learning to link COVID-19-related news sentiment with S&P 500 performance. Other studies explore unconventional sentiment sources, such as music sentiment (Can et al., 2023), weather conditions (Ma et al., 2023) and social media discussions (Long et al., 2023). Moreover, Ding et al. (2023) examined sentiment derived from technical trading strategies, finding that it predicts near-term returns but has an inverse effect in the long run. ...

Reference:

Time-frequency dynamics of public sentiment transmission in GCC equity markets: evidence from textual analysis

... Established around 1990, China's stock market features a unique dual-listing system [62]. In this study, dual-listed stocks refer to companies simultaneously listed on China's A-share market and the Hong Kong Stock Exchange (H-share). ...

Reference:

Factor-GAN: Enhancing stock price prediction and factor investment with Generative Adversarial Networks

... Notwithstanding the simplicity of a momentum strategy, the drawbacks of naïve price momentum are largely applicable to style momentum, namely increased portfolio turnover (trading costs), idiosyncratic volatility, market risk and crash risk (see Barroso & Santa-Clara, 2015;Daniel & Moskowitz, 2016;Grobys et al., 2018). ML offers a more advanced approach to estimating style rotation signals by considering multiple explanatory variables (features) and allows for the incorporation of non-linear dynamics between features and their dependent variables (labels), namely future style performance (see Galakis et al., 2021;Karatas & Hirsa, 2021;Ma et al., 2023;Neuhierl et al., 2023). ...

Reference:

Machine learning style rotation – evidence from the Johannesburg Stock Exchange

... Computer software, patents, copyrights, licences, customer lists, fishing permits and import limits are considered intangible assets under this definition. Liao et al. (2020) show that intangible assets pose challenges in their valuation and usability as collateral for financing. This results in significant financing constraints for firms with substantial intangible assets. ...

Reference:

How do intangible assets and financial constraints affect stock returns in Vietnam before and during the COVID‐19 pandemic?

... As noted above, the vast majority of studies, 19, were conducted in US markets, followed by Global with 5, see Figure 7. More recently, two studies have been dedicated to China by Ma et al. (2023) andLi et al. (2023), which have become seminal studies for this region. Figure 8 shows the databases used by the 25 studies. ...

Reference:

Factor Timing in Asset Management: A Literature Review