|
on Financial Markets |
| By: | Linying Lv |
| Abstract: | This paper documents new investment value in analyst reports. Analyst narratives embedded with large language models strongly forecast future stock returns, generating significant alpha beyond established analyst-based and fundamental-based factors. The return predictability arises primarily from reports that convey negative sentiment but forecast favorable long-term prospects, suggesting systematic market overreaction to near-term negative news. The effect is more pronounced for large, mature firms and for reports authored by skilled, experienced analysts. A Shapley value decomposition reveals that analysts' strategic outlook contributes the most to portfolio performance, especially forward-looking discussions on fundamentals. Beyond demonstrating untapped value in qualitative information, this paper illustrates the broader potential of artificial intelligence to augment, rather than replace, expert human judgment in financial markets. |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2502.20489 |
| By: | Yiming Ma; Kairong Xiao; Yao Zeng |
| Abstract: | Liquidity provision is often attributed to debt-issuing intermediaries like banks. We develop a unified theoretical framework and empirically show that mutual funds issuing demandable equity also provide an economically significant amount of liquidity by insuring against idiosyncratic liquidity shocks. Quantitatively, bond funds provide 12.5% of the liquidity that banks provide per dollar. Our model further shows that when equity values incorporate the liquidation cost from redemptions, as in swing pricing, liquidity provision is not necessarily reduced. This is because swing pricing may increase funds' capacity for holding illiquid assets without inducing panic runs. |
| JEL: | G2 G21 G23 G28 |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:33472 |
| By: | Leonardo Berti; Gjergji Kasneci |
| Abstract: | Stock Price Trend Prediction (SPTP) based on Limit Order Book (LOB) data is a fundamental challenge in financial markets. Despite advances in deep learning, existing models fail to generalize across different market conditions and struggle to reliably predict short-term trends. Surprisingly, by adapting a simple MLP-based architecture to LOB, we show that we surpass SoTA performance; thus, challenging the necessity of complex architectures. Unlike past work that shows robustness issues, we propose TLOB, a transformer-based model that uses a dual attention mechanism to capture spatial and temporal dependencies in LOB data. This allows it to adaptively focus on the market microstructure, making it particularly effective for longer-horizon predictions and volatile market conditions. We also introduce a new labeling method that improves on previous ones, removing the horizon bias. We evaluate TLOB's effectiveness using the established FI-2010 benchmark, which exceeds the state-of-the-art by an average of 3.7 F1-score(\%). Additionally, TLOB shows improvements on Tesla and Intel with a 1.3 and 7.7 increase in F1-score(\%), respectively. Additionally, we empirically show how stock price predictability has declined over time (-6.68 absolute points in F1-score(\%)), highlighting the growing market efficiencies. Predictability must be considered in relation to transaction costs, so we experimented with defining trends using an average spread, reflecting the primary transaction cost. The resulting performance deterioration underscores the complexity of translating trend classification into profitable trading strategies. We argue that our work provides new insights into the evolving landscape of stock price trend prediction and sets a strong foundation for future advancements in financial AI. We release the code at https://github.com/LeonardoBerti00/TLOB. |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2502.15757 |
| By: | Hamid Moradi-Kamali; Mohammad-Hossein Rajabi-Ghozlou; Mahdi Ghazavi; Ali Soltani; Amirreza Sattarzadeh; Reza Entezari-Maleki |
| Abstract: | Financial Sentiment Analysis (FSA) traditionally relies on human-annotated sentiment labels to infer investor sentiment and forecast market movements. However, inferring the potential market impact of words based on their human-perceived intentions is inherently challenging. We hypothesize that the historical market reactions to words, offer a more reliable indicator of their potential impact on markets than subjective sentiment interpretations by human annotators. To test this hypothesis, a market-derived labeling approach is proposed to assign tweet labels based on ensuing short-term price trends, enabling the language model to capture the relationship between textual signals and market dynamics directly. A domain-specific language model was fine-tuned on these labels, achieving up to an 11% improvement in short-term trend prediction accuracy over traditional sentiment-based benchmarks. Moreover, by incorporating market and temporal context through prompt-tuning, the proposed context-aware language model demonstrated an accuracy of 89.6% on a curated dataset of 227 impactful Bitcoin-related news events with significant market impacts. Aggregating daily tweet predictions into trading signals, our method outperformed traditional fusion models (which combine sentiment-based and price-based predictions). It challenged the assumption that sentiment-based signals are inferior to price-based predictions in forecasting market movements. Backtesting these signals across three distinct market regimes yielded robust Sharpe ratios of up to 5.07 in trending markets and 3.73 in neutral markets. Our findings demonstrate that language models can serve as effective short-term market predictors. This paradigm shift underscores the untapped capabilities of language models in financial decision-making and opens new avenues for market prediction applications. |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2502.14897 |
| By: | Yuan Liao; Xinjie Ma; Andreas Neuhierl; Linda Schilling |
| Abstract: | Machine learning in asset pricing typically predicts expected returns as point estimates, ignoring uncertainty. We develop new methods to construct forecast confidence intervals for expected returns obtained from neural networks. We show that neural network forecasts of expected returns share the same asymptotic distribution as classic nonparametric methods, enabling a closed-form expression for their standard errors. We also propose a computationally feasible bootstrap to obtain the asymptotic distribution. We incorporate these forecast confidence intervals into an uncertainty-averse investment framework. This provides an economic rationale for shrinkage implementations of portfolio selection. Empirically, our methods improve out-of-sample performance. |
| Date: | 2025–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2503.00549 |
| By: | Shicheng Zhou; Zizhou Zhang; Rong Zhang; Yuchen Yin; Chia Hong Chang; Qinyan Shen |
| Abstract: | This paper analyses the investment returns of three stock sectors, Manuf, Hitec, and Other, in the U.S. stock market, based on the Fama-French three-factor model, the Carhart four-factor model, and the Fama-French five-factor model, in order to test the validity of the Fama-French three-factor model, the Carhart four-factor model, and the Fama-French five-factor model for the three sectors of the market. French five-factor model for the three sectors of the market. Also, the LSTM model is used to explore the additional factors affecting stock returns. The empirical results show that the Fama-French five-factor model has better validity for the three segments of the market under study, and the LSTM model has the ability to capture the factors affecting the returns of certain industries, and can better regress and predict the stock returns of the relevant industries. Keywords- Fama-French model; Carhart model; Factor model; LSTM model. |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2502.05210 |
| By: | Steven J. Davis; Dingqian Liu; Xuguang Simon Sheng; Yan Wang |
| Abstract: | China’s stock market greatly outperformed other national markets during the first several months of the COVID-19 pandemic, and it did so even before it became evident that early containment efforts would flounder in the United States and many other countries. As to why, one view holds that aggressive monetary and credit easing propped up Chinese equity values. To assess this view, we consider several interventions that eased monetary and credit conditions in the first six months of 2020. Our analysis finds clear evidence that these interventions raised implied stock market volatility but little evidence that they influenced stock price levels. We also consider policy actions that restricted short selling, limited stock sales, and boosted stock purchases. These efforts to raise net equity demand were small in scale and highly time-limited, as we discuss, suggesting that any direct effects on stock prices were also modest. Neither our study nor other work known to us provides a ready explanation for the extraordinary performance of China’s stock market in the first half of 2020. This performance is even more striking in hindsight, given later developments in China’s economy and stock market. |
| JEL: | E52 E58 E65 G12 G18 |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:33485 |
| By: | Pawanesh Pawanesh; Charu Sharma; Niteesh Sahni |
| Abstract: | In financial networks, information does not always follow the shortest path between two nodes but may also take alternate routes. Communicability, a network measure, resolves this complexity and, in diffusion-like processes, provides a reliable measure of the ease with which information flows between nodes. As a result, communicability appears to be an important measure for detecting disturbances in connectivity within financial systems, similar to instability caused by periods of high volatility. This study investigates the evolution of communicability measures in the stock networks during periods of crises, showing how systemic shocks strengthen the pairwise interdependence between stocks in the financial market. In this study, the permutation test reveals that approximately 83.5 per cent of stock pairs were found to be statistically significant at the significance level of 0.001 and have an increase in the shortest communicability path length during the crisis than the normal days, indicating enhanced interdependence and heightened information flow in the market. Furthermore, we show that when employed as features in the classification model, the network shortest path-based measures, along with communicability measures, are able to accurately classify between the times periods of market stability and volatility. Additionally, our results show that the geometric measures perform better in terms of classification accuracy than topological measures. These findings provide important insights into market behaviour during times of increased volatility and advance our understanding of the financial market crisis. |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2502.08242 |