Weekly Research Recap
Latest research on investing and trading
Welcome to this week’s briefing, your curated roundup of the most actionable investing insights from the past seven days, drawing from academic research, industry reports, blogs, and social media, with links to everything.
Crypto
Volatility Transmission to Bitcoin: The Role of VIX Term Structure and Crypto Options Markets (Luo, Tsai, and Yen)
Bitcoin prices react instantly to volatility shocks from both crypto and equity markets. A 1-point rise in VIX cuts same-day Bitcoin returns by 0.68%, dwarfing Bitcoin’s own IV impact (-0.25%). The VIX term-structure slope dominates: Its coefficient (-0.77) is over twice the level effect. After Jan-2024 ETF approval, Bitcoin’s sensitivity to its own IV collapses, but VIX influence persists. Key takeaway: Watch the VIX curve, not crypto IV, to manage Bitcoin risk.
Equities
Interest Rates and Equity Valuations (Gormsen and Lazarus)
The authors decompose real rates into expected growth (survey forecasts), risk (VIX²), and a discounting residual, the portion left after regressing rates on growth and risk. Only this residual moves equity valuations, explaining 80% of cross-country valuation changes since 1990; in the U.S., just 35% of the rate decline passed through to stocks. Key takeaway: Don’t necessarily equate lower rates with higher stocks, only the discounting component truly matters.
Short Selling Around News in International Stock Markets (Gorbenko)
Short sellers across 38 countries reliably predict negative returns, but trading on negative news adds incremental alpha in only 6 markets. Outside the U.S., the evidence suggests most short-seller edge seems to arise from anticipating bad news via private information rather than superior public news processing. Key takeaway: Follow shorts as a signal of hidden information, news itself is rarely the alpha.
FX
Assessing cross-currency predictability in forex markets: Insights from limit order book data (Petrova, Vilhelmsson, and Norden)
Using limit order book data on five FX pairs, the authors test PCA, supervised PCA, LASSO, and random forests at 1-minute to 1-hour horizons. All models underperform a historical-average benchmark. Cross-currency features add little, and only order flow shows brief 1-minute predictability, pointing to short-lived microstructure effects, not durable alpha. Key takeaway: Any edge in modern FX order-book data is tiny and fleeting, don’t expect persistent, tradeable predictability.
Machine Learning & Large Language Models
Enhancing Asset Allocation and Portfolio Rebalancing Through Dynamic Theme Detection (Rubio)
Using NLP-driven product “themes” instead of static GICS sectors, the paper builds a dynamic thematic equity portfolio. In backtests, the strategy delivers 180.8% cumulative return vs 132.7% for SPY, albeit with higher volatility and deeper drawdowns (Sharpe is 0.97 vs. SPY 1.02). Key takeaway: AI-based theme classification can potentially uncover return drivers that traditional sectors miss.
Expectation Bias and Short-term Momentum (Gao, Ma, and Yuan)
Using ML to extract predictable analyst forecast errors (expectation bias), the paper finds a strong return signal: A long–short portfolio earns about 0.92% per month (11% annually) with a Sharpe ratio of 0.67. Expectation bias explains why 1-month momentum appears mainly in high-turnover stocks. Key takeaway: Short-term stock momentum comes from investors’ slow updating of their beliefs following news.
Large Language Models and Generative Factor Discovery in Crypto Markets (Sun, Wang, and Zhang)
The authors use GPT-5.2 to auto-generate crypto factors on BitMEX, producing alpha: The top long–short signal earns 658% annualized (Sharpe 4.13) in-sample and 808% (Sharpe 4.63) out of sample. A dynamic multi-factor portfolio returns 373% (Sharpe 3.21). Key takeaway: LLMs can uncover market-neutral crypto alpha, but strict controls on overfitting are needed.
Deep Reinforcement Learning for Optimal Portfolio Allocation: A Comparative Study with Mean-Variance Optimization (Sood, Papasotiriou, Vaiciulis, and Balch)
Using S&P 500 sector data, a deep reinforcement learning (DRL) allocator consistently beats mean–variance optimization: 12.1% vs 6.5% annual return and Sharpe 1.17 vs 0.68, with similar max drawdowns (-33%). Key takeaway: Optimizing directly for risk-adjusted returns via DRL can materially outperform classical MVO in real multi-asset allocation.
Portfolio Construction
Pairwise Dissimilarity and Risk-Seeking Portfolio Construction (Ryabinin)
The paper introduces a portfolio that overweights assets most statistically different from peers. In sector rotation, it beats equal-weighting by 28 to 121 bps/year with similar volatility and drawdowns (Sharpe 0.65–0.71); asset-allocation tests add 25–35 bps/year. Gains come from momentum capture and priced tail risk. Key takeaway: Letting dissimilarity in assets’ return distributions drive weights naturally loads the portfolio on trends and tail premia.
Blogs
Moneyball: Finding Undervalued Pairs Using Unconventional Metrics (Robot Wealth)
Crisis Alpha with Positive Carry and -0.5 Correlation to SPY (QuantSeeker)
Combining Calendar Strategies into the Trading Portfolio (Quantpedia)
Why Static Portfolios Fail When Risk Regimes Change (CFA Institute)
Time Series Foundation Models for Financial Markets: Kronos and the Rise of Pre-Trained Market Models (Jonathan Kinlay)
Podcasts
Peak Bubble? Why Markets Feel Different in 2026 ft. Mark Rzepczynski & Alan Dunne (Top Traders Unplugged)
What Druckenmiller Style Investing Gets Wrong - Alfonso Pecatiello on Edge in Macro Trading (Odds on Open)
Hendrik Bessembinder: Constant Leverage & Measuring Investor Outcomes (Rational Reminder)
Social Media & Industry Research
Making portfolio optimisation understandable for humans (Raul Leote de Carvalho)
Why Bitcoin Is Not the New Gold (Campbell Harvey)
Last Week’s Most Popular Links
Improving Performance with Fast Alphas; A Tactical Overlay for Intraday Trend Trading (Zarattini and Pagani)
Systematic Trend-Following with Adaptive Portfolio Construction: Enhancing Risk-Adjusted Alpha in Cryptocurrency Markets (Bui and Nguyen)
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.