Weekly Recap
Latest research on investing and trading
Welcome to this week's roundup of the latest investing research! Below is a carefully curated selection of highlights from the past week, with each title linking directly to its source for further reading.
Thank you for reading and don’t forget to hit the like button!
Commodities
Tail Risk Premium in the Crude Oil Market (Li and Li)
While the variance risk premium has been widely studied in financial markets, this paper finds that the option-implied tail risk premium is a stronger predictor of crude oil futures returns. Short-term tail risks signal lower returns in the next month but higher returns two months later. A trading strategy based on these insights outperforms others, offering better risk-adjusted returns for investors.
Tapping the Value of Futures Data (Till and Eagleeye)
The paper explores how traders use futures prices to understand oil market fundamentals. It highlights big data’s potential in commodities, challenges from data gaps like unavailable or incomplete fundamental data, and how price relationships reveal supply-demand shifts. It also covers key indicators, geopolitical impacts, and the need for greater transparency.
Crypto
Distrust and Cryptocurrency Price Deviations (Tang, You, and Zhong)
Why does Bitcoin trade at a premium in some countries? This paper finds that a lack of confidence in local governments is a key driver. Government corruption, economic turmoil, and restrictions on capital movement increase local demand for Bitcoin and Ethereum, pushing their prices above global levels. The effect is strongest in countries with weak institutional trust and tight financial controls.
Currencies
Cross-currency basis futures: description (Henrard)
CME Group recently launched the EUR/USD Cross-Currency Basis Futures contract. This paper explains its fundamentals, pricing, and how it connects to existing FX and interest rate markets, including futures and OTC derivatives.
Equity
Rethinking the Stock-Bond Correlation (Roncalli)
The stock-bond correlation has fluctuated significantly over time, both in magnitude and sign. This paper examines the factors driving these changes, linking them to factors such as inflation, economic growth, and monetary policy. It also discusses the implications for portfolio diversification and asset allocation.
The Aggregated Equity Risk Premium (Azevedo, Riedersberger, and Velikov)
The paper presents a new approach to predicting stock market returns. It first estimates expected returns for individual stocks using deep learning and then aggregates these predictions to forecast overall market returns. This method significantly outperforms traditional models, leading to higher investment returns and Sharpe ratios, particularly during economic downturns.
The Usefulness of EBITDA (Elfrink, Gee, Hills, and Whipple)
How useful is EBITDA for investors? The authors find that EBITDA is effective at predicting future operating cash flows, especially when firms have different capital structures. However, it is less useful for forecasting future earnings and free cash flow. Using adjusted EBITDA, excluding additional non-ITDA items like special charges and stock-based compensation, improves its predictive power for cash flows.
Reverse Timing of Insider Trading (Ma and Jiang)
Do corporate executives manipulate the timing and content of public information to benefit their preplanned trades under SEC Rule 10b5-1? This paper suggests they strategically release corporate disclosures before such trades, influencing stock prices in their favor. These findings raise concerns about the effectiveness of insider trading regulations.
Fear, Not Risk, Explains Asset Pricing (Arnott and McQuarrie)
Traditional risk-based explanations for stock market returns and anomalies have in many cases failed. Instead, the paper suggests that investor fear, both the fear of losing money and the fear of missing out, better explains market behavior. This alternative model helps make sense of long-standing anomalies and inconsistent returns observed throughout history.
Fixed Income
The Cross-Section of Corporate Bond Returns (Baltussen, Muskens, and Verwijmeren)
The authors study the drivers of U.S. corporate bond returns, addressing data issues and transaction costs. They find that a five-factor model, including market, bond value, equity momentum, short maturity, and accruals, best explains the cross-section of returns.
Machine Learning and Large Language Models
The LLM Quant Revolution: From ChatGPT to Wall Street (Mann)
The paper provides a review of Large Language Models in finance and their various use cases, including investment research, portfolio optimization, and trading strategy development. Specific examples discussed include earnings report analysis and automating backtesting code generation. The author advocates a multi-model approach, where, for example, GPT-4 is used for idea generation, while BloombergGPT processes corporate documents for deeper financial insights.
Big Data and Machine Learning in ESG Research (Li)
This paper explores how machine learning and big data enable researchers to analyze complex ESG issues using large-scale datasets. It reviews techniques, from basic text analysis to advanced deep learning, showing how they help uncover insights on corporate culture, climate risks, and other ESG factors that were previously difficult to measure.
Simulating the Survey of Professional Forecasters (Hansen, Horton, Kazinnik, Puzzello, and Zarifhonarvar)
Traditional economic surveys are costly, infrequent, and limited in flexibility, making it difficult to capture timely insights. This paper explores how large language models can simulate economic forecasts made by professional forecasters. By using AI-generated forecasters modeled on real participant data, the study finds that AI predictions closely match human forecasts and sometimes outperform them, particularly for long-term projections. This approach could make forecasting more frequent, cost-effective, and flexible while providing new insights into expert predictions.
Inferring Trade Directions in Options via Machine Learning (Gau, Tang, Zhou, and Zhou)
The paper discusses the rise of retail trading and price improvement auctions in options markets and how they have made it harder to correctly classify trades as buys or sells. A machine-learning model, GS-LASSO, is introduced that significantly improves trade classification accuracy.
An End-To-End LLM Enhanced Trading System (Zhou and Mehra)
Several studies suggest that investor sentiment can provide valuable trading signals. This paper introduces a trading system that leverages FinGPT to extract sentiment signals from financial news and social media in real-time. By integrating these signals with technical indicators, the system significantly outperforms traditional strategies.
Options
0DTE Index Options and Market Volatility: How Large is Their Impact? (Amaya, Garcia-Ares, Pearson, and Vasquez)
Market makers' gamma positions influence market volatility: when they are long gamma, their hedging trades tend to stabilize prices, but when they are short gamma, their rebalancing can amplify market moves. This paper finds that periods of negative gamma can sharply increase volatility, by up to 3.3 percentage points daily.
From Options to Fractions: The Effects of Fractional Trading on the Options Market (Crego and Garcia-Ares)
Before fractional trading, retail investors often used affordable call options to gain exposure to high-priced stocks. With fractional trading, they can now buy small portions of shares directly, reducing their reliance on options. The paper shows that this shift leads to wider bid-ask spreads and higher trading costs in the options market, impacting liquidity and market makers.
Portfolio Construction
Portfolio Construction Evolution: From Models to Machine Learning (Mann)
Portfolio optimization has evolved from traditional models to advanced machine learning techniques. This paper explores two key approaches: one groups investment signals by themes, creating diversified, market-neutral portfolios, while the other consolidates all signals into a single optimized portfolio using mean-variance optimization. It compares their strengths, examines implementation frameworks, and provides practical guidance with Python implementations.
Risk Parity, Maximum Diversification and Conviction (Fusai and Mignacca)
Investors often seek to balance risk and return in their portfolios. This paper connects two popular approaches, Risk Parity and Maximum Diversification, by introducing the concept of conviction, which measures uncertainty in an asset’s risk contribution. The authors show that incorporating conviction naturally leads to a more diversified portfolio, providing a new framework for portfolio optimization.
Retirement Planning
How the 4% Rule Would Have Failed in the 1960s: Reflections on the Folly of Fixed Rate Withdrawals (McQuarrie)
The 4% rule, introduced by William Bengen in 1994, was originally tested on U.S. market data from 1926 to 1992. This paper argues that it would have failed investors at multiple points in history due to high inflation and weak market returns. It critiques fixed withdrawal rates as unreliable and advocates for flexible strategies or alternatives like TIPS ladders for more sustainable retirement income.
Blogs
Turn-of-the-Month Strategies: Do They Still Work? (QuantSeeker)
Coding Trend Factor (Quantitativo)
How much should we get paid for skew risk? Not as much as you think! (Rob Carver)
Taming Excessive “Timing Luck” in TAA by Tranching Strategies (Allocate Smartly)
GitHub
finmarketpy - Python library for backtesting
Medium
How to Find Seasonality Patterns in Time Series (Mezzini)
Identify High Value Stocks with the Piotroski F-Score (Velasquez)
How to Build an Algorithmic Trading System with Python (DataScience Nexus)
Podcasts
I beat Vegas betting on the NBA, put 160% of my net worth into Bitcoin and now own a pro soccer team (My First Million)
How to Choose an Asset Allocation (Rational Reminder)
Why Quants Might Be Your Best Bet ft. Nick Baltas (Top Traders Unplugged)
Last Week’s Most Popular Links
Market Making in Crypto (Stoikov, Zhuang, Chen, Zhang, Wang, Li, and Shan)
Causal Network Representations in Factor Investing (Howard, Lohre, and Mudde)
Goal Alpha: A Polymarket and EPL Study (Puri)
Disclaimer: This newsletter is for informational and educational purposes only and should not be construed as investment advice. The author does not endorse any specific securities or investments mentioned. While information is gathered from sources believed to be reliable, there is no guarantee of its accuracy, completeness, or correctness.
This content does not offer 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 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 are the author's own interpretations of the findings and content. These summaries should not be considered as definitive or comprehensive representations of the original works. Readers are encouraged to refer to the original sources for complete and authoritative information.
This newsletter contains 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 these external sites or for that of subsequent links. Users click on these links at their own risk.