Welcome to this week's roundup of the latest investing research! Below is a carefully curated selection of highlights from the past week, with each title linking directly to its source for further reading.
Thank you for reading and don’t forget to hit the like button!
Betting
Goal Alpha: A Polymarket and EPL Study (Puri)
The paper explores Premier League betting on the Polymarket platform. It finds that betting odds are not perfectly efficient, with low-probability outcomes tending to be undervalued and high-probability ones overpriced, contrary to the classical long-shot bias. Additionally, odds adjust slowly after goals, suggesting that delayed market reactions could create short-term trading opportunities.
Crypto
Market Making in Crypto (Stoikov, Zhuang, Chen, Zhang, Wang, Li, and Shan)
The authors develop a market-making strategy for crypto perpetual contracts using one-minute candlestick data from 30 cryptocurrencies. They implement and test their approach on the Hummingbot platform, introducing a new trading signal, "Bar Portion", which outperforms a standard MACD-based strategy in backtesting and live trading.
Equities
What Moves Prices? The Dynamics of Fundamentals and Returns (Girardi and Schlag)
The paper studies the relative importance of discount rate variations versus expected cash flow growth in driving stock prices. It finds that while stock price fluctuations have historically been dominated by changes in discount rates, expectations about fundamental growth, especially earnings growth, have become increasingly important in recent decades, particularly for earnings-based valuations.
Value Investing: Integrating Theory and Practice (Lee)
Value investing focuses on buying stocks that are undervalued relative to their fundamentals. Legendary investors like Warren Buffett emphasize both quality and price. The paper highlights how accounting-based valuation helps identify strong businesses trading below their intrinsic worth, showing that those combining quality and undervaluation tend to outperform.
Monetary Policy and Hedge Funds' Reaching for Beta (Abdi, Chen, and Wu)
Hedge funds adjust their stock market exposure in response to monetary policy. They buy high-beta stocks after an interest rate increase and buy low-beta stocks after an interest rate decrease. This means they take on more market risk when monetary policy tightens and reduce risk when policy eases. Skilled hedge funds use this strategy effectively, generating higher returns by interpreting Fed announcements as signals about future economic conditions.
Factor Investing
Causal Network Representations in Factor Investing (Howard, Lohre, and Mudde)
The authors explore how causal discovery algorithms can enhance factor investing, overcoming the limitations of conventional correlation-driven approaches. By applying causal network models to the S&P 500, the authors show that these methods can improve stock groupings, factor selection, and market timing. While promising, the approach remains computationally demanding and requires further work.
Fixed Income
Forecasting Corporate Bond Index Returns (Cao, Song, Yang, and Zhan)
Corporate bond index returns can be predicted using company financial data and economic indicators. Evaluating different methods, Partial Least Squares is found to be the most effective. Investors using these forecasts can enhance bond investment strategies and improve returns.
Long-Term Debt and Short-Term Rates (De Stefani and Mano)
The paper explores how fixed-rate mortgages (FRMs) influence central banks' ability to manage the economy through interest rate changes. When rates are lowered, more people opt for FRMs, reducing the impact of future rate hikes. This weakens monetary policy effectiveness over time, making economic stabilization harder.
Mind the Inflation Swap (Luber and Press)
The authors document a consistent overestimation of inflation expectations when relying on inflation swap rates, especially during periods of high uncertainty. This distortion affects key financial metrics like real interest rates and the inflation risk premium. A method that corrects for this bias using inflation options is introduced, providing more reliable inflation estimates.
Machine Learning and Large Language Models
Can deep reinforcement learning beat 1/N (Kruthof and Muller)
Can deep reinforcement learning outperform a simple equal-weighted investment strategy? By testing a specific algorithm on over 300 years of market data, the study finds that while deep reinforcement learning can identify market timing signals, it fails to consistently outperform the equal-weighted approach, especially when transaction costs are considered.
The Application of Machine Learning in Finance: Situation and Challenges (Zhang)
Machine learning (ML) is transforming finance by improving areas such as risk management and forecasting. This survey paper reviews current use cases of ML in Finance and discusses the pros and cons of common methods and the potential integration of areas such as quantum computing and blockchain.
Bitcoin Price Direction Forecasting and Market Variables (Kim, Jo, Choi, and Jang)
The authors introduce a Bitcoin forecasting model that combines convolutional neural networks with long short-term memory networks. Interestingly, they train separate models for predicting price increases and decreases, then merge their outputs. Key predictive features include U.S. interest rates, the U.S. dollar index, and commodity prices. The model significantly outperforms a buy-and-hold strategy.
Large Language Models in Finance: Reasoning (Noguer I Alonso)
Large language models can process text well but struggle with complex financial reasoning. This paper explores ways to enhance their decision-making, such as structured thinking and external data use. It highlights applications such as investment analysis and risk management while addressing challenges related to accuracy and transparency.
Sentiment Management: AI-based Evidence from Earnings Guidance (Berkovitch, Israeli, and Kasznik)
The paper explores how companies adjust the emotional tone of their earnings guidance to influence investors. It finds that firms highlight positive aspects in titles and early sections while downplaying negative information. Investors react to this sentiment, incorporating it into stock prices and trading activity, suggesting strategic communication affects market behavior.
The Natural Language of Finance (Hoberg and Manela)
This is a comprehensive survey paper on the role of natural language processing in financial research, highlighting a range of practical applications, the evolution of techniques from early word-count methods to today’s generative AI models, and ongoing challenges such as addressing biases in financial texts and improving causal inference.
Mutual Funds
Mutual Fund Investors and the Economic Cost of Seeking Alpha (Friesen and Nguyen)
Mutual fund investors have shifted from chasing past returns to prioritizing risk-adjusted performance and alpha over time. However, this shift has not improved outcomes, investors tend to buy funds after periods of strong performance, leading to poor timing decisions that reduce returns. Only the most skilled investors benefit from chasing alpha.
Options
December Effect in Option Returns (Choy, Wei, and Zhang)
The authors document that delta-hedged option returns are significantly lower than usual during the month of December. Options tend to be overpriced in early December as traders overlook the holiday-induced drop in volatility. The anomaly pertains to both stock and S&P 500 index options and selling straddles in December generates significant returns.
The Fed and the Wall Street Put (Harren, Kilic, and Zhang)
The paper explores how financial firms trade S&P 500 options on days when the Federal Reserve announces policy decisions. It finds that proprietary trading firms sell options early in the day, well before the announcement. These trades often anticipate looser monetary policy and lower option prices, and the paper suggests that some firms may have advance knowledge of the Fed’s moves.
Blogs
Exploring Stock-Bond Correlations (QuantSeeker)
Cross-country equity risk allocation with statistical learning (Macrosynergy)
Seasonality Patterns in the Crisis Hedge Portfolios (QuantPedia)
GitHub
Awesome-Quant-Machine-Learning-Trading
Medium
Candlestick Pattern Recognition with YOLO (Velasquez)
Algorithmic Trading with Python: A Simple Strategy That Beats Buy and Hold (Algo Insights)
An easy way to beat $SPY with these two simple indicators (Filip)
Podcasts
Should you EVER Take Partial Profits? (Line Your Own Pockets)
Thao Tran – Market Making Illiquid, Non Fungible Assets (Flirting with Models)
Stock Leaders and Timeless Strategies Still Effective Today · Ross Haber (Chat With Traders)
Essential Lessons from History's Best Traders | Jack Schwager (Excess Returns)
Last Week’s Most Popular Links
Developing Cryptocurrency Trading Strategy Based on Autoencoder-CNN-GANs Algorithms (Hu, Yu, Zhang, Zheng, Liu, and Zhou)
Conditional short-term trend signals (Macrosynergy)
Disclaimer: This newsletter is for informational and educational purposes only and should not be construed as investment advice. The author does not endorse any specific securities or investments mentioned. While information is gathered from sources believed to be reliable, there is no guarantee of its accuracy, completeness, or correctness.
This content does not offer personalized financial, legal, or investment advice and may not be suitable for your individual circumstances. Investing carries risks, and past performance does not guarantee future results. The author and affiliates may hold positions in securities discussed without prior notification.
The author is not affiliated with, sponsored by, or endorsed by any of the companies, organizations, or entities mentioned in this newsletter. Any references to specific companies or entities are for informational purposes only.
The brief summaries and descriptions of research papers and articles provided in this newsletter are the author's own interpretations of the findings and content. These summaries should not be considered as definitive or comprehensive representations of the original works. Readers are encouraged to refer to the original sources for complete and authoritative information.
This newsletter contains links to external websites and resources. The inclusion of these links does not imply endorsement of the content, products, services, or views expressed on these third-party sites. The author is not responsible for the accuracy, legality, or content of these external sites or for that of subsequent links. Users click on these links at their own risk.