|
on Financial Markets |
| By: | Junjie Guo |
| Abstract: | This paper provides an empirical study explores the application of deep learning algorithms-Multilayer Perceptron (MLP), Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and Transformer-in constructing long-short stock portfolios. Two datasets comprising randomly selected stocks from the S&P500 and NASDAQ indices, each spanning a decade of daily data, are utilized. The models predict daily stock returns based on historical features such as past returns, Relative Strength Index (RSI), trading volume, and volatility. Portfolios are dynamically adjusted by longing stocks with positive predicted returns and shorting those with negative predictions, with equal asset weights. Performance is evaluated over a two-year testing period, focusing on return, Sharpe ratio, and maximum drawdown metrics. The results demonstrate the efficacy of deep learning models in enhancing long-short stock portfolio performance. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.13555 |
| By: | Jue Xiao; Tingting Deng; Shuochen Bi |
| Abstract: | In recent fast-paced financial markets, investors constantly seek ways to gain an edge and make informed decisions. Although achieving perfect accuracy in stock price predictions remains elusive, artificial intelligence (AI) advancements have significantly enhanced our ability to analyze historical data and identify potential trends. This paper takes AI driven stock price trend prediction as the core research, makes a model training data set of famous Tesla cars from 2015 to 2024, and compares LSTM, GRU, and Transformer Models. The analysis is more consistent with the model of stock trend prediction, and the experimental results show that the accuracy of the LSTM model is 94%. These methods ultimately allow investors to make more informed decisions and gain a clearer insight into market behaviors. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.05790 |
| By: | Matteo Bonato (Department of Economics and Econometrics, University of Johannesburg, Auckland Park, South Africa; IPAG Business School, 184 Boulevard Saint-Germain, 75006 Paris, France); Rangan Gupta (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa); Christian Pierdzioch (Department of Economics, Helmut Schmidt University, Holstenhofweg 85, P.O.B. 700822, 22008 Hamburg, Germany) |
| Abstract: | Recent global economic and political events have made clear that shortages are a key factor driving macroeconomic and financial market developments. Against this backdrop, we studied the forecasting value of shortages for monthly U.S. stock market realized variance (RV) at the aggregate and sectoral level using data spanning the period 1900-2024 and 1926-2023 (for most sectors), respectively. To this end, we considered linear and non-linear statistical learning estimators. When we used linear estimators (OLS and shrinkage estimators), we did not find evidence that aggregate and disaggregate shortage indexes have predictive value for subsequent market or sectoral RVs. In contrast, when we used random forests, a nonlin- ear nonparametric estimator, we detected that aggregate and disaggregate shortage indexes improve forecast accuracy of market and sectoral RVs after controlling for realized moments (realized leverage, realized skewness, realized kurtosis, realized tail risks). We then decomposed RV into a high, medium, and low frequency component and found that the shortages indexes are correlated mainly with the medium and low frequencies of RV. |
| Keywords: | Shortages, Stock market, Realized volatility, Statistical learning, Forecasting |
| JEL: | C22 C53 E23 G10 G17 |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:pre:wpaper:202450 |
| By: | Sorouralsadat Fatemi; Yuheng Hu |
| Abstract: | Financial trading has been a challenging task, as it requires the integration of vast amounts of data from various modalities. Traditional deep learning and reinforcement learning methods require large training data and often involve encoding various data types into numerical formats for model input, which limits the explainability of model behavior. Recently, LLM-based agents have demonstrated remarkable advancements in handling multi-modal data, enabling them to execute complex, multi-step decision-making tasks while providing insights into their thought processes. This research introduces a multi-modal multi-agent system designed specifically for financial trading tasks. Our framework employs a team of specialized LLM-based agents, each adept at processing and interpreting various forms of financial data, such as textual news reports, candlestick charts, and trading signal charts. A key feature of our approach is the integration of a reflection module, which conducts analyses of historical trading signals and their outcomes. This reflective process is instrumental in enhancing the decision-making capabilities of the system for future trading scenarios. Furthermore, the ablation studies indicate that the visual reflection module plays a crucial role in enhancing the decision-making capabilities of our framework. |
| Date: | 2024–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.08899 |
| By: | Szymon Lis |
| Abstract: | This study conducted a comprehensive review of 71 papers published between 2000 and 2021 that employed various measures of investor sentiment to model returns. The analysis indicates that higher complexity of sentiment measures and models improves the coefficient of determination. However, there was insufficient evidence to support that models incorporating more complex sentiment measures have better predictive power than those employing simpler proxies. Additionally, the significance of sentiment varies based on the asset and time period being analyzed, suggesting that the consensus relying on the BW index as a sentiment measure may be subject to change. |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.13180 |
| By: | Pongsak Luangaram; Yuthana Sethapramote; Kannika Thampanishvong; Gazi Salah Uddin |
| Abstract: | Understanding the impact of climate risks on financial stability is crucial for ensuring the resilience of banking sectors, particularly in economies exposed to climate change. This paper investigates how transition and physical risks influence systemic risk in Thailand’s banking sector. Transition risks are analyzed using the Fama-French multi-factor asset pricing model to estimate the risk premium of brown industries relative to green industries, termed Brown-minus-Green (BMG). Physical risks are assessed using the Standardized Precipitation Evapotranspiration Index (SPEI), an indicator of flood and drought conditions. Systemic risk at the bank level is measured using conditional value-at-risk (CoVaR). Panel regressions are employed to examine the relationship between climate risks and systemic risk. The results reveal that transition risks, as captured by the BMG factor, significantly heighten systemic risk among Thai banks, emphasizing their critical role in financial vulnerabilities. Additionally, physical risks, particularly those associated with flood exposure, create substantial challenges for bank portfolios. These findings highlight the importance of integrating transition and physical risk indicators into regulatory monitoring frameworks to enhance financial stability. Furthermore, Thai commercial banks can apply these insights to conduct climate stress tests and develop strategies for managing climate-related risks more effectively. |
| Keywords: | climate risk; Systemic risk; Thailand; Banking sector; BMG; SPEI; CoVar |
| JEL: | C58 G12 G21 Q54 |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:pui:dpaper:224 |
| By: | Linying Lv |
| Abstract: | I examine the value of information from sell-side analysts by analyzing a large corpus of their written reports. Using embeddings from state-of-the-art large language models, I show that textual information in analyst reports explains 10.19% of contemporaneous stock returns out-of-sample, a value that is economically more significant than quantitative forecasts. I then perform a Shapley value decomposition to assess how much each topic within the reports contributes to explaining stock returns. The results show that analysts' income statement analyses account for more than half of the reports' explanatory power. Expressing these findings in economic terms, I estimate that early acquisition of analysts' reports can yield significant profits. Analysts' information value peeks in the first week following earnings announcements, highlighting their vital role in interpreting new financial data. |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.13813 |
| By: | Ananya Unnikrishnan |
| Abstract: | Reinforcement learning (RL) has emerged as a transformative approach for financial trading, enabling dynamic strategy optimization in complex markets. This study explores the integration of sentiment analysis, derived from large language models (LLMs), into RL frameworks to enhance trading performance. Experiments were conducted on single-stock trading with Apple Inc. (AAPL) and portfolio trading with the ING Corporate Leaders Trust Series B (LEXCX). The sentiment-enhanced RL models demonstrated superior net worth and cumulative profit compared to RL models without sentiment and, in the portfolio experiment, outperformed the actual LEXCX portfolio's buy-and-hold strategy. These results highlight the potential of incorporating qualitative market signals to improve decision-making, bridging the gap between quantitative and qualitative approaches in financial trading. |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.11059 |
| By: | Maria Teresa Medeiros Garcia; Tiago Miguel Batista Raimundo |
| Abstract: | This paper examines how player transfers influence the stock prices of publicly traded football clubs through an event study approach. The analysis focus on five prominent European teams—Manchester United, Juventus, Borussia Dortmund, Olympique Lyon, and Ajax—focusing on 230 player transactions that occurred between 2018 and 2023. The study assesses abnormal returns (AR) and cumulative abnormal returns (CAR) within a 10-day event window, encompassing five days prior to and following the announcements of transfers. Findings indicate that high-value transfers typically result in positive abnormal returns, which reflect investor optimism regarding the new player's potential impact on the team's success. In contrast, sales and loans of players tend to elicit negative reactions from the market, indicating concerns about possible adverse effects on team performance. These results support the Efficient Market Hypothesis by demonstrating that stock prices quickly adjust to new information such as player transfers. This research adds to the expanding literature at the intersection of sports events and financial markets, providing valuable insights for clubs operating in capital markets and investors aiming to understand the dynamics of football markets. |
| Keywords: | Event Studies; Football Transfers; Abnormal Returns; Stock Market; Efficient Market Hypothesis. |
| JEL: | G14 L83 M41 |
| Date: | 2024–12 |
| URL: | https://d.repec.org/n?u=RePEc:ise:remwps:wp03612024 |
| By: | Athanasios Geromichalos; Lucas Herrenbrueck; Changhyun Lee; Sukjoon Lee (Department of Economics, University of California Davis) |
| Abstract: | We build on recent developments in the theory of money and liquidity to provide a qualitative and quantitative explanation for the well-known TIPS illiquidity vis-`a-vis noninflation- protected Treasuries. Our model does not assume exogenous differences between the markets where the two assets are traded or the investors who hold them; instead, an asset’s liquidity is endogenous and depends on the trading and market entry decisions of investors. The model offers a powerful amplification mechanism that operates upon only one difference between TIPS and Treasuries: the far greater supply of the latter. We also quantify how much the Treasury leaves on the table by issuing securities that are not as highly valued by the market as nominal Treasuries. However, the model does not necessarily imply that the Treasury should phase out TIPS, as some have suggested, only that it should seek to reduce segmentation and ensure that TIPS trade alongside nominal Treasuries in consolidated secondary markets. |
| Keywords: | Monetary-search models, OTC markets, Endogenous liquidity, Treasury securities, Treasury Inflation-Protected Securities (TIPS) |
| JEL: | E31 E43 E52 G12 |
| Date: | 2024–12–02 |
| URL: | https://d.repec.org/n?u=RePEc:cda:wpaper:364 |
| By: | Yuval Boneh; Ethan Jones |
| Abstract: | This paper delves into the spectrum of credit risks associated with decentralized stablecoin issuance, ranging from overcollateralized lending to business-to-business credit. It examines the mechanisms, risks, and mitigation strategies at each layer, highlighting the potential for scaling decentralized stablecoins while ensuring systemic health. |
| Date: | 2024–11 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2411.13762 |