Weekly Research Recap
Latest research on investing and trading
Here’s this week’s Tuesday roundup, a selection of the most practical investing ideas from recent academic studies, industry research, blogs, and insightful discussions across social platforms, with links included throughout.
Asset Allocation
Regime-Aware Asset Allocation with Dual-Regime Signals and Regime-Dependent Asset Selection (Luo and Mulvey)
The authors combine global and asset-specific regime forecasts to dynamically rotate across equities, bonds, REITs, commodities, gold, and cash. Their dual-regime allocation framework achieved an out-of-sample Sharpe ratio of 1.43 from 2000 to 2025. Key takeaway: The edge may come less from forecasting market direction and more from identifying the assets most likely to thrive in each market regime.
Crypto
Order flow and cryptocurrency returns (Anastasopoulos, Gradojevic, Liu, Maynard, and Tsiakas)
A measure of global cryptocurrency order flow, the balance of buying and selling pressure across exchanges and currencies, strongly predicts future returns. Weekly order-flow portfolios achieved a Sharpe ratio of 1.9, while machine learning models with daily-rebalanced long-short portfolios achieved Sharpe ratios above 3.5. Key takeaway: In crypto, order flow appears to be a stronger predictor of future returns than fundamentals.
Equities
In good and in bad times? The Relation between anomaly returns and market states (Muller and Preissler)
Across 133 equity anomalies in 56 markets, abnormal returns were 70% higher during unfavorable market states, with roughly 75% of the performance gain coming from the short side. Key takeaway: Anomaly returns appear strongest when market conditions are poor, consistent with mispricing and limits-to-arbitrage playing a major role in their persistence.
A new decomposition approach to modeling financial returns: Conditioning sign on magnitude (Brou and Luger)
Rather than predicting returns in the standard way, this paper suggests decomposing returns into direction and magnitude. By forecasting the probability of gains and losses separately from the magnitude of market moves, the authors uncover nonlinear patterns that standard predictive regressions miss, thereby improving out-of-sample market timing. Key takeaway: Volatility contains information not only about risk but also about the direction of future returns.
Factor themes and the principal-component structure of factor momentum (Kim and Ryu)
Decomposing factor momentum into within- and across-theme components, the authors find that most of the variation in factor momentum comes from shifts between themes such as Value, Quality, Momentum, and Low Risk. The across-theme component is also more robust to trading costs and closely linked to the common drivers of factor returns. Key takeaway: Factor momentum may be driven more by rotation between factor themes than by selection within individual factors.
Measures of Accounting Quality and the Cross-Section of Stock Returns (Wang)
Many classic accounting-quality signals have lost their predictive power. This paper finds the edge has migrated elsewhere: Beneish’s M-score and a machine-learning Accounting Quality Score continue to predict future returns, while traditional accrual measures largely fail. Key takeaway: The accounting-quality premium is probably better measured today by misstatement-detection signals than by traditional accrual measures.
When Does Insider Trading Work? Evidence from Weekly Timing and Investor Attention (Ma, Cheng, Ding, Fu, and Zhou)
Insider purchases cluster on Fridays, while sales cluster on Mondays. Using 2.6 million insider trades, this paper finds that Friday purchases and Monday sales are the most informative insider signals, especially when investor attention is high. Key takeaway: Insider-trading signals depend on timing as much as information.
Futures
Does the Nikkei Quanto Spread Contain a Monetizable Risk Premium? (Tan)
This paper shows that a naive implementation of the so-called Nikkei quanto premium is unlikely to survive transaction costs. However, the author identifies a small but persistent edge linked to the market’s tendency to extrapolate recent stock–currency covariance and suggests alternative implementations. Key takeaway: The alpha appears to come from mispriced covariance.
Machine Learning and Large Language Models
International corporate bond returns: Uncovering predictability using machine learning (Li, Lu, Qi, and Zhou)
This paper applies machine learning to 24,000 corporate bonds across 86 countries and finds meaningful return predictability in both U.S. and international markets. The most important signals differ by region: Momentum dominates in the U.S., while downside risk and illiquidity matter more abroad. Long-short portfolios achieved Sharpe ratios of roughly 1 to 1.6. Key takeaway: The drivers of corporate bond returns vary across regions, suggesting that successful bond signals may need to be market-specific.
Technical indicators and the cross-section of corporate bond returns in a machine learning era (Chin, Guo, Lin, and Mei)
Technical indicators derived from prices, yields, and trading volume are highly informative for forecasting cross-sectional corporate bond returns. Surprisingly, machine learning adds little beyond simple linear models. Key takeaway: In corporate bonds, market-based signals appear to contain more predictive information than traditional bond characteristics, and complexity doesn’t necessarily beat simplicity.
Portfolio Construction
Selection versus diversification in noisy alpha environments (Goto and Yamada)
Investors are sometimes told to discard signals that fail stringent statistical tests. This paper finds that doing so can hurt portfolio performance. Portfolios built from a broader set of signals outperform portfolios restricted to only strongly statistically significant predictors. Key takeaway: For portfolio construction, diversification across signals can outweigh strict significance-based signal selection.
Blogs
Is There Alpha in the COT Report? (QuantSeeker)
FIFA* World Cup (*Fitting and Forecasting Actual data) Portfolio Optimisation competition with real returns (Rob Carver)
Enhancing bond index returns with systematic FX hedging (Macrosynergy)
A Hidden Trade Around SpaceX IPO? (Concretum Group)
Shock Accounting (John H. Cochrane)
Podcasts
When Trend Following Meets Equities ft. Eric Crittenden & Andrew Beer (Top Traders Unplugged)
The 3 Timeless Rules Shared by 100 Years of Market Wizards · Jack Schwager & George Coyle (Chat With Traders)
Neeraj Khemlani and Matt Ankrum, CFA: The Hunt for 100-Bagger Stocks (Enterprising Investor)
Social Media & Industry Research
The Factor Zoo Has Hundreds of Animals — But Only a Handful of Species (Alpha Architect)
Dividend Timing and Global Dividend Premium (Alpha Architect)
The whirlwind is upon us (Acadian Asset Management)
Something Has Got to Give (Eventually) (Man Group)
Last Week’s Most Popular Links
Rejoicing, Regret and Stock Returns – US and International Evidence (So and Zhang)
Price Path Continuity and the Cross-Section of Cryptocurrency Returns (Kim)
Crowded Anomalies over the Business Cycle (Jung)
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.
Paid subscriptions may not be available in all jurisdictions and may change without notice.