Weekly Recap
Latest research on investing and trading
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!
Crypto
Token Economics (Gregory and Mini)
Blockchain networks rely on digital tokens to function effectively without a central authority. This paper explores how tokens are designed, distributed, and used to encourage participation and maintain stability. It covers several mechanisms like Binance Coin’s (BNB) token burning, which reduces supply to increase value, MakerDAO’s DAI minting system, where users deposit collateral to generate stablecoins, and Aave’s governance tokens, which allow users to vote on protocol changes.
Quantifying Cryptocurrency Unpredictability: A Comprehensive Study of Complexity and Forecasting (Puoti, Pittorino, and Roveri)
Cryptocurrency prices are notoriously volatile. This study examines their randomness and forecasting potential by comparing price patterns to well-known random processes. The findings suggest that cryptocurrencies behave much like Brownian motion, essentially random walks, meaning their price changes are highly unpredictable. Interestingly, simple forecasting methods performed just as well or better than advanced machine learning models, highlighting the challenge of finding meaningful patterns in crypto markets.
Currencies
How Credible is Hong Kong's Currency Peg? (Jermann, Wei, and Yue)
The paper examines the stability of Hong Kong’s currency peg to the U.S. dollar. While the peg has been in place since 1983, financial markets now assign a growing risk to its breakdown due to rising U.S. interest rates and a weakening Hong Kong economy. Using an asset-pricing model and HKD option prices, the authors estimate that market confidence in the peg has declined, with the risk of a break reaching levels last seen during the global financial crisis.
Equities
What are Asset Price Bubbles? A Survey on Definitions of Financial Bubbles (Baumann and Janischewski)
The authors explore how financial bubbles are defined across different disciplines, showing that no single definition dominates. While many economists describe bubbles as asset prices rising above their fundamental values, other definitions focus on price movements, speculation, or market dynamics. The lack of consensus makes it challenging to detect bubbles in real time or regulate them effectively.
Informative Price Pressure (Arif, Da, and Lin)
Before major Federal Reserve (FOMC) meetings, investors hedge their stock positions, creating temporary price pressures. These movements reflect informed investors’ expectations about future returns, but in reverse: a market drop before the meeting signals optimism for the long run. This pattern reliably predicts stock performance for up to two years. Investors can use pre-FOMC price changes as a potential indicator of long-term market trends.
Cross-predictability of returns: An equilibrium perspective (Konermann and Meinerding)
Returns on certain stocks can help predict future returns on others due to economic connections between companies. Instead of relying on information delays or market frictions, this study finds that when cash flow shocks ripple through the economy, they create return patterns that investors can use. By tracking these connections, traders can build long-short strategies that systematically exploit return predictability.
Does the Factor Zoo Pay Off? A Portfolio View on Mispricing and the Limited Gains from New Anomalies (Lassance and Martin-Utrera)
Many stock characteristics appear to predict returns, but practical barriers like trading costs and short-selling constraints reduce their real-world usefulness. This research examines whether a portfolio designed to exploit these inefficiencies can improve risk-adjusted returns. The results show that while such a portfolio can be beneficial, gains have not improved with newer anomalies since the 1980s. Investors should be cautious about relying on newly discovered factors, as many fail to provide lasting benefits.
Out-of-Sample Alphas Post-Publication (Goncalves, Loudis, and Ogden)
The paper examines why anomaly-based investment strategies, which seem profitable in research studies, often fail to generate strong returns once they become widely known. A key challenge is that investors don’t know how much to allocate to these strategies in real-time, making it difficult to achieve the expected gains. While combining multiple anomalies can improve returns, investors aren't aggressive enough to eliminate profits. This suggests that individual anomalies may not be as useful to investors post-publication.
Finding the Missing Momentum in China (Peng, Wang, and Xu)
Momentum strategies, which rely on the tendency of past winners to continue outperforming, work well in many markets but seem weak in China. This study finds that retail investors play a key role in dampening momentum effects by reacting to non-fundamental factors like price anchors and lottery-like stocks. However, after removing these distractions, a strong momentum signal emerges. Investors can enhance momentum strategies in China by focusing on fundamental trends and avoiding stocks heavily influenced by retail speculation.
The Predictive Power of Inter-trade Durations: Return Reversals and Momentum (Saba)
The time gap between trades provides useful clues about price movements. When trading slows down, prices are less likely to experience sudden reversals and more likely to continue in the same direction. This effect is stronger in calm markets and when institutional investors are active, as their trades tend to be more informed. Investors can use trade duration as a signal: longer gaps suggest stronger trends, while shorter gaps hint at possible reversals.
Machine Learning and Large Language Models
ChatGPT and Deepseek: Can They Predict the Stock Market and Macroeconomy? (Chen, Tang, Zhou, and Zhu)
ChatGPT can effectively identify good and bad news from financial headlines and predict market returns up to six months ahead, outperforming other large language models like DeepSeek. Investor underreaction to good news, especially during economic downturns, is a key reason why return predictability exists.
Do Trades and Holdings of Market Participants Contain Information About Stocks? A Machine-Learning Approach (DeMiguel, Guo, Sang, and Zhang)
Many investors analyze trades and holdings of different market participants to uncover signals about future stock returns. Traditional studies focus on specific investor types, but this paper uses machine learning to capture interactions across multiple investor groups, including hedge funds, banks, and other institutions. A strategy based on both trades (changes in holdings) and existing holdings generates strong returns, particularly in smaller, less-followed stocks where market inefficiencies are more pronounced.
The DeepFake and its Impact on Trading Signals (Shamo)
Markets rely on accurate information, but deepfake technology can generate fake news, videos, and announcements that manipulate investor sentiment and trading signals. These fabrications mislead traders and algorithms, fueling volatility, misinformation-driven trades, and loss of trust. Investors should verify sources and leverage fraud detection tools to safeguard their decisions.
Market Microstructure
To Make, or to Take, That Is the Question: Impact of LOB Mechanics on Natural Trading Strategies (Albers, Cucuringu, Howison, and Shestopaloff)
Many traders assume that placing passive orders in an order book (maker orders) is efficient due to lower fees, but this strategy often leads to unprofitable trades. Using data from Bitcoin perpetual contracts on Binance, the paper shows that maker orders tend to fill when prices are about to move unfavorably, while aggressive (taker) orders struggle to overcome exchange fees. However, the authors identify specific reversal setups where maker orders can achieve both high fill probability and favorable price movement, making them profitable.
Volatility
Asset Co-movement, Investor Sentiment, and Market Volatility: An Analysis of the VIX and MOVE Index (Liu and Clarkson)
When market uncertainty rises, stocks and bonds from the same company can move together in ways that investors might not expect. This paper finds that when stock market volatility (VIX) and bond market volatility (MOVE) increase together, they help predict future stock and bond returns, as well as how these assets move relative to each other.
Option Implied Timing with Ambiguity and Risk (Lu)
The traditional belief that higher risk leads to higher returns has been questioned by anomalies like the low-risk effect. This research explores whether investors can improve performance by adjusting their exposure based on ambiguity and volatility signals derived from option prices. The findings suggest that using option-implied information leads to better risk-adjusted returns than traditional volatility-based timing. Investors can apply these insights by incorporating option-based signals to enhance performance.
Blogs
Exploiting Short-Term Mean Reversion Between Stocks and Bonds (QuantSeeker)
Classifying credit markets with macro factors (Macrosynergy)
What’s the chance that a market effect is real? Monte Carlo permutation tests in Excel (RobotWealth)
Modelling the yield curve of US government treasuries (Open Source Quant)
GitHub
toraniko - A multi-factor equity risk model for quantitative trading
AlphaPy - Python AutoML for Trading Systems and Sports Betting
Cvxportfolio - Portfolio optimization and back-testing
Medium
Explaining Transformers as Simple as Possible through a Small Language Model (Punnen)
12 Python Libraries for Free Market Data That Everyone Should Know (DataScience Nexus)
How to Distill a LLM: Step-by-step (Leo)
Podcasts
A $33 Billion Value Manager Who Has Actually Outperformed | Scott McBride (Excess Returns)
Michael Mauboussin: The One Job of an Equity Investor (Rational Reminder)
Position Sizing Is Sexy (Line Your Own Pockets)
Last Week’s Most Popular Links
The LLM Quant Revolution: From ChatGPT to Wall Street (Mann)
Fear, Not Risk, Explains Asset Pricing (Arnott and McQuarrie)
An End-To-End LLM Enhanced Trading System (Zhou and Mehra)
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.