Weekly Recap
Latest research on investing and trading
It’s time for yet another roundup of the latest investing research from the past week! Below is a carefully curated selection of highlights, with each title linking directly to its source for further reading.
Thank you for reading and don’t forget to hit the like button.
Commodities
Commodity dependence and optimal asset allocation (Dequiedt, Gomes, Pukthuanthong, and Williams)
Investors often assume adding commodities to a portfolio enhances diversification. However, this paper finds that the benefits depend on a country’s economic structure. In nations heavily reliant on commodity exports, local stocks and bonds move in sync with commodity prices, limiting diversification advantages. Conversely, investors in less commodity-dependent economies can improve portfolio performance by including commodities. This highlights the importance of tailoring asset allocation to a country's risk exposure.
Raising the Bar - Commodity Price Predictability via Iterated Combinations (Ma, Zhou, and Wang)
The authors introduce a new forecasting method for commodity returns that iteratively blends multiple linear prediction models, improving accuracy over traditional approaches. By incorporating macroeconomic variables and technical indicators, it outperforms both standard models and machine learning techniques. This approach leads to better risk-adjusted returns, as evidenced by higher Sharpe ratios in asset allocation tests.
Crypto
Spot-Futures Manipulations in Cryptocurrency Markets (Wang and Zhang)
Cryptocurrency markets often lack the regulations that prevent manipulation in traditional finance. This paper uncovers a scheme where traders with large holdings inflate crypto spot prices, drawing in retail investors, then profit by shorting futures as prices fall. Using Binance data, it finds clear patterns of such manipulations, leading to losses for retail traders.
Solana and the SOL Market (Overdahl and Lewis)
Solana addresses the scalability and speed limitations of earlier blockchains like Bitcoin and Ethereum. This paper provides a detailed overview of Solana, covering its market structure, Proof of History consensus mechanism, economic model, liquidity, and decentralization.
Volatility Clustering in Bitcoin (Borrego Roldan)
Bitcoin experiences periods of stability followed by bursts of volatility, a common pattern in financial markets. This paper confirms that Bitcoin’s volatility clusters, meaning high or low volatility tends to persist. Investors can use this persistence to anticipate market conditions, adjust risk exposure, and refine hedging strategies, particularly when managing extreme price movements.
Currencies
Common Factors in Currency Characteristics (Dauber and Umlandt)
The paper explores how common patterns in currency characteristics can better explain currency returns than traditional models. Instead of analyzing individual factors like carry or momentum separately, it identifies a few latent factors that capture the broader structure of currency markets. The main finding is that a two-factor model, one similar to the carry factor and another acting as a hedge against carry crash risk, effectively explains the cross-section of currency returns.
(Un)Expected Political Outcomes and Currency Markets (Filippou, Li, and Liu)
Political surprises, such as unexpected election outcomes, can significantly impact currency markets. This research finds that after such events, currencies often appreciate, particularly when the results catch markets off guard. The effect is strongest when driven by economic uncertainty and geopolitical instability.
Equities
Resurrecting the value effect: The role of technology stocks (Lee)
Over the past 20 years, the value premium has faded, puzzling investors. This paper attributes much of the decline to the rise of technology firms, whose high valuations distort traditional value classifications, leading to an underestimation of the value premium. By sorting tech and non-tech stocks separately, the paper demonstrates that the value effect remains strong when accounting for industry differences.
The 10-year Expected Return of the S&P 500 in early 2025 (Philips and Kobor)
Many investors use the CAPE ratio, which smooths earnings over a decade, to estimate future stock returns. However, this paper shows that a simpler approach, using just one year of earnings and filtering out noisy data, can be even more effective. By refining how earnings are measured and incorporating non-linear relationships, the model improves return forecasts, estimating long-term stock returns at around 4.2% per year.
Dissecting Momentum in China (Liu, Tan, Xu, Yuan, and Zhu)
Chinese stocks don’t follow the usual momentum pattern seen in other markets. Traditional momentum signals are distorted by retail investors who push prices up on news days and institutional investors who correct these overreactions on non-news days. Instead of past returns, a news-based momentum signal, such as the percentage of recent positive news, better captures underreaction to fundamentals while avoiding retail-driven price swings. This approach produces more reliable trading signals in China’s A-share market.
Factor Investing
Multiple Hypothesis Testing, Empirical Asset Pricing, and Factor Investing (Shi)
Many investment strategies rely on identifying patterns in historical data, but testing hundreds of factors increases the risk of finding patterns that appear significant just by chance, the so-called multiple testing problem. Traditional methods often fail to account for this, leading to false discoveries. This paper provides an extensive review of multiple testing techniques applied to factor investing.
Behavioral Finance and Factor Investing (Shi)
The efficient market hypothesis suggests that stock prices reflect all available information, but behavioral finance shows that investor biases can lead to mispricing. Psychological tendencies like overconfidence and overreaction create patterns such as momentum and reversal, and even skilled investors struggle to correct these inefficiencies due to risks and costs. This paper explores these behavioral biases in-depth, summarizing their impact on investor decision-making and discussing their implications for factor investing and asset pricing.
Fixed Income
Interest Rate Instruments and Market Conventions Guide - Post LIBOR edition (Henrard)
With LIBOR’s phaseout, financial markets have adopted new benchmarks and conventions for interest rate products. This guide breaks down key instruments, their structure, and the rules that govern them, covering overnight rates, benchmark indices, clearing practices, and swaps.
Hedge Funds
What is the Future of Alternative Investing? (Ennis)
Many institutional investors have embraced alternative investments, like private equity and hedge funds, believing they provide diversification and strong returns. However, the paper argues that these investments come with high fees, often exceeding 3-4% per year while delivering returns comparable to traditional stock and bond portfolios. The author suggests that, over time, institutional investors will likely shift toward low-cost index funds as they recognize the inefficiencies of alternative investments.
Machine Learning and Large Language Models
ChatGPT and Deepseek: Can They Predict the Stock Market and Macroeconomy? (Chen, Tang, Zhou, and Zhu)
Several papers have studied the ability of large language models to classify news for stock return prediction. This paper tests ChatGPT and DeepSeek using Wall Street Journal news and finds that ChatGPT can extract valuable information that predicts stock market returns, while DeepSeek is much less effective. The predictability stems from investors’ slow reaction to positive news, especially in uncertain times.
A Deep Learning Framework for Medium-Term Covariance Forecasting in Multi-Asset Portfolios (Reis, Serra, and Gama)
Predicting asset return correlations is challenging, as traditional models like GARCH and shrinkage estimators struggle with changing market regimes. This research develops a deep learning framework that improves covariance matrix forecasting, demonstrating superior accuracy compared to a range of more traditional approaches.
(Re-)Imag(in)ing Price Trends in China Using Multi-Channel Grayscale Image CNN Models (Chen, Xue, Zhao, and Zhu)
A growing body of research applies Convolutional Neural Networks to stock price images for return prediction. This paper extends that approach to China's A-share market by converting stock data into grayscale images and analyzing them with deep learning models. The findings suggest that incorporating additional information, such as trading volume, past returns, and moving averages, enhances prediction accuracy and investment performance.
Options
The factor structure of short maturity options (Yoo)
Short-term index options, especially those expiring within days, have surged in popularity and trading volume. This study examines the factors driving their returns and finds that a two-factor model incorporating volatility and skewness risk explains the cross-section of option returns well.
Portfolio Construction
Causal Factor Analysis is a Necessary Condition for Investment Efficiency (Lopez de Prado, Lipton, and Zoonekynd)
Equity factors have traditionally been defined based on correlations rather than causation. As a result, many factor strategies have underperformed out-of-sample, leading to investor disappointment. This paper argues that using a causal approach to factor modeling is essential to minimize model misspecification and improve investment efficiency.
Trading
Refining ETF Asset Momentum Strategy (Pauchlyova and Vojtko)
This paper enhances an ETF momentum strategy across asset classes by incorporating a correlation filter and selective shorting. The approach goes long the four top-performing ETFs while shorting one ETF with a 30% weight only when short-term correlations are high, leading to better performance than standard momentum strategies.
Blogs
Trading the Fed: The Pre-FOMC Drift is Alive (QuantSeeker)
Using Inflation Data for Systematic Gold and Treasury Investment Strategies (QuantPedia)
How Far Back Should You Backtest? (Portfolio123)
GitHub
FinQuant - Financial portfolio management, analysis and optimisation
Medium
Implied Volatility Analysis for Insights on Market Sentiment (Velasquez)
HMA Crossover Strategy: A Data-Driven Approach to Trading (Kridtapon P.)
Can YOLO Predict Stock Market Ups and Downs? (Velasquez)
Podcasts
Rob Carver - The Comprehensive Guide to a Diversified Futures Strategy (The Algorithmic Advantage)
Jim Bianco on What a 'Mar-a-Lago Accord' Could Mean for the Economy (Odd Lots)
Investors Made These Costly Mistakes in 2024 - Here is How You Can Avoid Them | Larry Swedroe (Excess Returns)
Last Week’s Most Popular Links
The Predictive Power of Inter-trade Durations: Return Reversals and Momentum (Saba)
toraniko - A multi-factor equity risk model for quantitative trading
Informative Price Pressure (Arif, Da, and Lin)
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.