Weekly Research Recap
Latest research on investing and trading
Plenty of interesting research this week. Here’s a concise recap of the most actionable investing insights from the past seven days, covering academic papers, industry reports, blogs, and social media, with direct links throughout.
Asset Allocation
A Causal Regime-Risk-Stability Framework for Global Tactical Asset Allocation (Bailer and Alber)
The authors present a GTAA framework combining regime detection, a forward-looking Risk Index, and an optional stability overlay. Trading a diversified institutional asset-class universe, GTAA delivers 7.1% p.a. at 9% volatility, while GTAA (Stable) cuts volatility to 6.8%, earns 5.5% p.a., and achieves the best drawdown and Sortino profile. Key takeaway: Reducing exposure when market conditions deteriorate results in smoother returns and smaller drawdowns without compromising long-term performance.
Crypto and Digital Assets
Sector Structure in Digital Asset Returns (Mao, Yang, and Caporin)
Digital assets show a clear sector structure, but not through differing market betas. Using 425 tokens (2020–2023), the paper finds that idiosyncratic sector shocks drive sector return differences. Sector factors are weakly correlated (<0.1), display short-lived momentum/reversal, and have minimal spillovers (≈3% connectedness). Key takeaway: Crypto sector rotation exploits sector-specific information, not beta timing.
Equities
Salience Theory and Risk Anomalies (Guo and Li)
Using U.S. stocks, the paper shows that the risk–return relation depends on return salience (returns that stand out as unusually large relative to the market). High-beta stocks underperform after salient upside returns (about −0.8% per month) but outperform after salient downside returns (about +0.5% per month). Retail attention and trading amplify this mispricing. Key takeaway: The beta premium is conditional, not unconditional.
MAX on Steroids: A New Measure of Investor Attraction to Lottery Stocks (Bali, Ince, Ozsoylev)
The paper shows that the classic MAX anomaly largely reflects persistent mispricing tied to equity issuance. By double sorting first on market beta and then on MAX, the authors construct MAXβ, isolating idiosyncratic extreme returns. A value-weighted long–short MAXβ strategy earns about −0.8% per month. Key takeaway: A beta-neutral MAX factor better captures lottery-like demand and delivers a cleaner measure of mispricing.
R&D Alpha: Investment Intensity and Long-Term Stock Returns (Sehgal)
High R&D intensity predicts higher future stock returns in U.S. large caps. Sorting S&P 500 firms by R&D-to-revenue yields a 3.7% annual high–minus–low premium over 1995–2025, with a FF5 alpha of 4.4%. An investable top-20 R&D strategy earns 7.5% annual excess return vs. SPY after costs from 2001–2025, with a Sharpe near 1.0. Key takeaway: Systematically tilting toward innovation-intensive firms delivers a simple, implementable source of equity alpha.
ECBetas in Equity Strategy Returns (McBride, Sarno, and Zinna)
Using ECB announcement-window OIS surprises, the paper shows a clear risk–return tradeoff across 153 euro-area equity factors. Factors with low or negative interest-rate betas earn 5–6% annualized with a Sharpe of 0.5 to 0.6, driven mainly by press-release shocks. Key takeaway: Exposure to monetary policy shocks is a systematic, priced risk, and equity factors that suffer when rates rise earn higher expected returns as compensation.
Machine Learning and Large Language Models
Does Large Language Models have Market Timing Ability? Evidence from Global Markets (Tang, Xie, and Zhu)
Using daily data from 30 equity markets (2000–2024), the paper evaluates LLMs as pure market timers: The model observes only the past 20 daily returns and predicts the sign of the next day’s market return. Performance is weak with no consistent timing gains. Key takeaway: Market timing based solely on past returns is not a reliable strength of LLMs.
Using daily data from 2020–2024, the paper shows that combining CNN’s Fear & Greed Index with VIX significantly improves next-day S&P 500 direction forecasts. A simple Decision Tree achieves 68.8% accuracy versus 57.6% using VIX alone, with sentiment changes driving most of the predictive power. Key takeaway: Fear & Greed adds short-horizon predictability, especially in low-to-moderate VIX regimes.
Regret-Driven Portfolios: LLM-Guided Smart Clustering for Optimal Allocation (Abro and Jaleel)
The paper tests an adaptive allocation rule that updates weights from recent performance, limits rebalancing in extreme sentiment regimes, and adds AI-guided hedges. On sector ETFs (2011–2024), it beats SPY by 69% cumulatively, more than doubles the Sharpe (1.0 vs. 0.47), and cuts max drawdown from 34% to 16%. Key takeaway: Simple adaptation plus downside control materially improves risk-adjusted returns.
Options
Stock Trading Variability and Option Returns (Wiest)
Options on stocks with volatile trading volume earn lower future delta-hedged returns. Sorting by turnover variability delivers a 1.1% monthly long–short spread. The effect is driven by upside volume spikes and reflects higher implied volatility required by market makers for hedging and jump risk. Key takeaway: Variability in stock turnover may be a useful signal for identifying options with elevated implied volatility and lower future returns.
Regime Models
A Piecewise-Linear Model For Market Regime Identification (Steiner)
The paper introduces a methodological approach to market regime identification based on detecting changes in price trends. By fitting connected linear segments and choosing the number of breaks via a complexity–fit tradeoff, regime slopes are interpreted as bull, flat, or bear phases. Applied to the Swiss Leaders Index, the model delivers a gain in explanatory power. Key takeaway: Piecewise trend models can be useful for diagnosing market regimes.
Blogs
Stock selection with macro factors: the case for simple neural networks (Macrosynergy)
Why Tight Stop-Losses Often Hurt Investors — and What Robust Capital Growth Really Requires (CFA Institute)
Who Is the Counterparty to the Pro-Cyclical Investors (Quantpedia)
Podcasts
Investor Insights #02: A Conversation With Quant Investor Yuval Taylor (Skull Sessions and Maj Soueidan)
Why 20% of Hedge Funds Fail After One Year - Claudia Quintela on Why Managers Need Business Sense (Odds on Open)
Building an Inflation-Proof Portfolio ft. Yoav Git (Top Traders Unplugged)
This Market May Be More Speculative Than the Dot-Com Bubble (Why No One Cares) - Richard Bernstein (Meb Faber Show)
Social Media & Industry Research
Is Value Investing Dead? (Alpha Architect)
Trifecta: A Fundamental Revolution in Indexing (Research Affiliates)
How To Correctly Reason About A Market Maker (SystematicLongShort)
2026 through a Systematic Lens: What to Expect When You’re Not Expecting (Man Group)
The Macro Implications of the AI Capex Boom (Bridgewater)
Moontower on Gamma (Kris Abdelmessih)
Last Week’s Most Popular Links
Regime-Aware Universal Portfolio (Vlasiuk and Smirnov)
Prediction markets and lotteries (and my one simple trick for winning the lottery) (Rob Carver)
Episodic Factor Pricing (Li, Yuan, and Zhou)
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.