Weekly Research Recap
Latest research on investing and trading
As always, this week’s recap curates the most actionable investing insights from the past seven days across academic research, industry reports, blogs, and social media, complete with direct links.
Crypto
Crypto Currency Returns (Jheng et al.)
This paper examines classic crypto studies using over 17,000 coins through 2025 and finds that their results still hold. Crypto remains a separate, high-risk asset, and adding it to a traditional portfolio lifts the Sharpe from 0.92 to 1.23. User-based valuation and crash-risk models continue to flag major bubbles, including 2020–21 and the 2024 ETF boom. Key takeaway: Crypto is now an institutional asset class, but it still behaves like a high-return, high-crash asset.
Good versus Bad COVOL in Cryptocurrency Markets: A Measure of Asymmetric Common Volatility (Pham, Han, Nguyen, Pham, and Do)
Crypto crashes are driven by market-wide “bad” volatility. By using their constructed relative COVOL Index (good minus bad common volatility), a simple market timing strategy, holding crypto only when upside volatility dominates, boosts cumulative returns to 142% vs 46% for buy-and-hold and raises Sharpe and Sortino by about 40 to 50%. Key takeaway: Market-wide asymmetric volatility can be used to market-time crypto.
Equities
Tail risk exposure and the cross section of expected stock returns (Nicolas)
Stocks that crash with the market seem to earn a premium, but most of that effect is just high correlation. When correlation is stripped out, true crash risk is only priced among low-correlation stocks. A double-sorted portfolio that first selects low-correlation stocks and then targets high crash exposure earns 6.8% per year, with a t-stat of 4.5 and a Sortino ratio of 4.15, outperforming single-sort portfolios. Key takeaway: Taking on crash risk only pays a risk premium in low-correlation stocks.
Machine Learning & Large Language Models
Expected Investment and Stock Returns: A Machine Learning Approach (Tian)
The paper evaluates multiple ML models and shows they predict firm investment far better than linear regressions. Expected investment is driven mainly by sales growth, momentum, and Tobin’s Q. A long–short strategy on high versus low expected investment firms earns 0.56% per month, with high-investment firms also exhibiting stronger profitability and growth. Key takeaway: ML-based forecasts of expected corporate investment uncover a tradable return premium for investors.
A Unified Framework for Anomalies based on Daily Returns (Cakici, Fieberg, Neszveda, Bianchi, and Zaremba)
The authors feed the last 21 daily returns into an elastic-net model that separates the timing of returns from their magnitudes. Timing dominates: Very recent moves drive short-term reversals far more than older, more extreme moves. The long–short factor earns 1.57% per month pre-costs (Sharpe 1.23), or 2.26% with equal-weighting (Sharpe 2.39), and remains profitable after reasonable trading costs. Key takeaway: Short-term reversal profits are mostly driven by the most recent price moves.
The Limits of Complexity: Why Feature Engineering Beats Deep Learning in Investor Flow Prediction (Kang)
Using 2.8 million trades across 2,439 Korean stocks, the paper shows that simple order-flow signals clearly outperform deep learning: A linear momentum strategy earns a Sharpe of 1.30 and +272.6% total return, while an ICA-Wavelet-LSTM framework delivers a Sharpe of close to zero. Key takeaway: In low signal-to-noise markets, smart feature engineering beats complex algorithms.
Cross-Market Alpha: Testing Short-Term Trading Factors in the U.S. Market via Double-Selection LASSO (Du, Walter, and Ulrich)
Testing short-term trading signals from China’s Alpha191 library on S&P 500 stocks from 2002 to 2022, the paper finds that 17 signals survive after controlling for 151 U.S. factors, driven by price–volume pressure, short-term reversals, and volatility asymmetry. Key takeaway: Fast, behavior-based signals built for the Chinese A-share market still carry predictive power in U.S. equities.
Hard to Process: Atypical Firms and the Cross-Section of Expected Stock Returns (Weibels)
Using an autoencoder to measure how unusual a firm’s financial profile is, the paper shows that stocks with hard-to-interpret profiles are systematically overpriced and then underperform. A low-minus-high ATYP portfolio earns +1.47% per month (equal-weighted) and +0.82% (value-weighted) before costs. Key takeaway: Screen out weird, hard-to-value stocks; they tend to be overpriced and deliver weak future returns.
Machine Learning, Classification Algorithm and Cross Section of Stock/Bond Returns (Chin)
Using a classification ML setup (predicting top- vs bottom-decile winners) dramatically improves bond return predictability compared to predicting returns. Long–short bond portfolios earn about 1.6% per month with a Sharpe of 0.6 to 0.8, beating regression models on returns and drawdowns. Stock results are mixed. Key takeaway: For predicting bond returns, rank-based ML classification models beat return forecasts, i.e., predict who wins and loses, not by how much.
Mutual Funds
Read (between) the lines (van den Berg, Jansen, Neefjes, Tetereva, and Voigt)
Managers who write with confidence in their shareholder letters earn higher future returns and lower risk. A long-only portfolio that holds the top funds ranked by language signals achieves a higher Sharpe ratio than portfolios built using standard fund metrics alone. Key takeaway: Clear, confident manager communication is an investable signal for improved risk-adjusted returns.
Options
Extracting Forward Equity Return Expectations Using Derivatives (Clark, Lu, and Tian)
Using index and VIX futures and options, the authors extract a term structure of expected stock returns. They find strong crisis-driven mean reversion: One-month-ahead equity premia exceed 20% annualized in downturns versus 4–7% in expansions. A timing strategy based on this signal delivers meaningful value. Key takeaway: Option-implied return expectations offer a way to time equity exposure, especially after large sell-offs.
Blogs
How dollar invoicing and dollar debt shape FX risk premia (Macrosynergy)
Much Ado About Variance (Robot Wealth)
How Multibaggers Really Happen (Quantseeker)
What Earnings Explain, and What They Don’t: Insights from 150 Years of Market Data (CFA Institute)
Are markets that are good for trend good just because they have also gone up a lot, or because carry, or... (Rob Carver)
Podcasts
Club Conversation with Aswath Damodoran, the Dean of Valuation (MicroCapClub)
Moritz Heiden & Moritz Seibert – Trend-Following Spreads (Flirting with Models)
This Ex-Poker Pro Built a Hedge Fund by Betting Against Beta – David Orr on Asymmetric Bets (Odds on Open)
World’s BEST Traders Expose The Top Trading Lessons for Profitability in 2026 (Words of Rizdom)
Why “Skin in the Game” Can Be Dangerous for Traders - Andreas Clenow (Odds on Open)
Social Media & Industry Research
DeepSeek Founder Liang’s Funds Surge 57% as China Quants Boom (Bloomberg)
Introduction To Practical Portfolio Optimization (SystematicLongShort)
‘Captain Condor’ Wipeout Offers Harsh Lesson in Managing Risk (Bloomberg)
Last Week’s Most Popular Links
Optimal Sharpe Ratio Portfolios and the Role of Perfect Foresight into Returns, Volatilities, and Correlation (Roy, Motson, and Thomas)
The Volatility You Can’t See (Concretum Group)
Gold and Silver Shine During the Day (Kelly)
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.