Time for another round of great investing research. Below is a curated list of last week’s highlights, each linked to the original source for easy access.
Appreciate your continued support! If you’re finding value in these posts, feel free to like and subscribe if you haven’t already.
Bonds
Cross-Asset Trend Spillover: A Novel Factor for Corporate Bond Returns (Fieberg, Liedtke, Schlag, and Zaremba)
Many studies use firm fundamentals or past bond prices to forecast corporate bond returns. This paper takes a different approach by relying solely on the issuing firm's stock price data. Using over 30 technical indicators in an elastic net model, the authors uncover strong predictive power that outperforms traditional models and remains robust across thousands of tests. Stock-based signals thus serve as valuable complements to standard bond return predictors.
Geopolitical skewness risk and bond returns (Jiang, Kang, and Tian)
Past research has shown that exposure to geopolitical risk is priced and associated with a risk premium. This paper introduces a new way of measuring such risks by constructing a refined skewness-based indicator using the country-specific components of geopolitical risk indices. The indicator is found to be a strong predictor of U.S. Treasury bond returns and provides meaningful improvements in asset allocation performance for investors.
Equities
Market-State Dependent Momentum Strategies: An Empirical Examination of Anomalies in Asset Pricing (Rodon Comas)
Momentum strategies often struggle in certain market environments, such as periods of high volatility or broad market declines. This paper shows that switching between momentum and reversal strategies based on the market's past two-year performance leads to more consistent returns. By aligning with the prevailing regime, following trends in strong markets and trading reversals in weak ones, the results suggest investors can lower drawdowns and boost risk-adjusted performance.
Market Signals from Social Media (Cookson, Lu, Mullins, and Niessner)
Sentiment from news and social media is known to have predictive power. This paper separates sentiment from attention and builds daily indexes from millions of posts on StockTwits, Twitter, and Seeking Alpha. Both indexes predict future returns negatively, but for different reasons: high sentiment follows gains and precedes reversals, while high attention follows losses and signals continued decline. A market timing strategy based on these signals delivers meaningful Sharpe ratios.
The EV Shakeout (Arnott, Cornell, Henslee, and Verghese)
Investors rushed to back electric vehicle startups, hoping they’d all become the next Tesla. The paper explains how fierce competition, high costs, and overly optimistic expectations led to massive losses as many of these companies failed to deliver. Only a few survived, and even Tesla’s dominance appears uncertain. The authors suggest investors focus on fundamentals rather than getting swept up in excitement and unrealistic valuations.
Machine Learning and Large Language Models
Foreign Signal Radar (Jiao)
Most machine learning papers on predicting U.S. stock returns use features based on U.S. data. Instead, this paper trains stock-specific machine learning models solely on foreign stock and index returns to forecast daily returns for S&P 500 firms. The approach identifies predictive foreign signals for U.S. stocks, generating double-digit returns net of costs in daily rebalanced portfolios.
Textual Regression for Realized Volatility: A Model for Long-Term Forecasting (Parvini and Assa)
Agricultural markets are often driven by external shocks like weather events. This paper forecasts volatility in agricultural futures and finds that news text contains valuable predictive information. The authors evaluate the forecasting power of text embeddings from various language models and show that LLaMA-based models, in particular, deliver strong results. These models significantly outperform both standard HAR benchmarks and sentiment-score-based approaches. For investors, this suggests that incorporating rich textual information into volatility models can meaningfully improve forecasts.
Predicting Financial Market Stress with Machine Learning (Aldasoro, Hördahl, Schrimpf, and Zhu)
Financial markets can suddenly become unstable, triggering tail risks that ripple across economies. Traditional tools often miss early warning signs in specific markets. This paper shows that machine learning models, trained on a rich set of market stress indicators, more accurately predict extreme events in FX and fixed-income markets. Models like random forest and XGBoost outperform standard methods, helping investors better anticipate and manage risk.
Large Language Models in Equity Markets: Applications, Techniques, and Insights (Jadhav and Mirza)
This paper reviews recent academic literature on how large language models (LLMs) are being applied to areas such as trading, prediction, and sentiment analysis. By covering 84 studies, it offers a practical overview of current capabilities, limitations, and emerging trends in LLM-driven equity strategies.
Artificial Intelligence and Machine Learning in Corporate Finance (Hornuf and Schaefer)
This is a comprehensive handbook chapter, providing a broad overview of how machine learning, artificial intelligence, and large language models are being used in corporate finance research and practice.
The More, the Better? Predicting Stock Returns with Local and Global Data (Cakici and Zaremba)
Does using global data instead of local data improve return predictions? By training elastic net models across 45 markets, the paper finds that global models offer little advantage over local ones when forecasting cross-sectional returns within a country. The only meaningful gains appear in smaller or more volatile markets with limited data. For most investors, local models are often simpler, cheaper, and equally effective.
Options
Expected Return or Mispricing? Evidence from the Term Structure of Option-implied Disagreement (Fan, Han, and Lu)
Investors often hold conflicting views on where markets are headed, especially in the short term. By analyzing option trading activity across different time horizons, the authors construct a term structure of investor disagreement. They find that near-term disagreement is larger, and that both the slope and curvature of this term structure positively predict future market returns. Investors could potentially use these signals to time their equity exposure.
Blogs
Quantamental economic surprise indicators: a primer (Macrosynergy)
Annual performance update returneth - year 11 (Rob Carver)
Weekly Research Insights (QuantSeeker)
GitHub
pysystemtrade - Systematic trading in Python by Rob Carver.
EliteQuant - A list of online resources for quantitative modeling, trading, and portfolio management.
hftbacktest - A high-frequency trading and market-making backtesting and trading bot in Python and Rust.
Medium
The HAR-X model for Volatility Trading. A new approach (Filip)
Finding Mispriced Stocks with a 6 Factor Model (Velasquez)
Adaptive Multi-Regime Sector Rotation Strategy (Filip)
Podcasts
WTF is this market? - with Vineer Bhansali of LongTail Alpha (The Derivative)
The Impact of Tariffs On Investing (The Economist Overseeing $6 Trillion Explains) (Meb Faber Show)
The Colossal Mistake | Richard Bernstein on the Risks of Tariffs and Passive Investing (Excess Returns)
Last Week’s Most Popular Links
Enhancing Trend-Following Strategies Using Machine Learning and Time Series Models (Chandrinos and Lagaros)
Automate Strategy Finding with LLM in Quant Investment (Kou, Yu, Luo, Peng, and Chen)
A Tale of Two Anomalies: The Implications of Investor Attention for Price and Earnings Momentum (Hou, Loh, Peng, and Xiong)
Disclaimer: This newsletter is for informational and educational purposes only and should not be construed as investment advice. The author does not endorse or recommend any specific securities or investments. While information is gathered from sources believed to be reliable, there is no guarantee of its accuracy, completeness, or correctness.
This content does not constitute 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, and these holdings may change at any time 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 should not be considered definitive or comprehensive representations of the original works. Readers are encouraged to refer to the original sources for complete and authoritative information.
This newsletter may contain 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 external sites or for that of any subsequent links. Users access these links at their own risk.
The author assumes no liability for losses or damages arising from the use of this content. By accessing, reading, or using this newsletter, you acknowledge and agree to the terms outlined in this disclaimer.