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.
Behavioral Finance
How Costly are Trading Heuristics? (Han, He, and Weagley)
Retail investors often rely on simple decision-making shortcuts when picking stocks, but these habits can be costly. By analyzing decades of research and actual trading data, the paper finds that traders frequently use heuristics like chasing lottery-like stocks (betting on extreme past winners) or herding (following the crowd into popular stocks), leading to lower future returns. Institutional investors, however, use fewer heuristics and often benefit from them. The key takeaway: blindly following simple rules, like preferring easy-to-pronounce stock names or doubling down on past winners, usually results in poor investment outcomes.
Crypto
Derivative Arbitrage Strategies in Cryptocurrency Markets (Valery)
Cryptocurrency markets often show inconsistencies in pricing, creating arbitrage opportunities. This paper explores a strategy that exploits mispricings between standard cryptocurrency options and binary options from prediction markets. By carefully combining these instruments, investors can construct trades that guarantee no losses at expiration. Although such opportunities are rare, they are highly profitable when they arise. Overall, the study highlights the potential for systematic arbitrage in crypto derivatives.
Currencies
Media tone is a priced risk factor in currency markets (Heimonen, Lehkonen, and Pukthuanthong)
Media sentiment, derived from millions of news and social media articles, influences currency markets by predicting returns beyond traditional economic indicators. The study uses Refinitiv’s MarketPsych Index and finds that currencies with more positive media sentiment tend to appreciate, especially in harder-to-value emerging markets. This effect lasts for months, outperforming simple forecasting models. Hence, the results suggest investors can use media sentiment as a trading signal, particularly for currencies where fundamental valuation is uncertain.
Equities
Does Central Bank Tone Move Asset Prices? (Schmeling and Wagner)
Changes in how central banks communicate can influence financial markets beyond their actual policy decisions. Using a dictionary-based approach, this paper examines ECB press conferences to measure shifts in tone and their effect on investor sentiment. A more positive tone leads to rising stock prices, lower credit spreads, and reduced market volatility, while also pushing interest rates higher.
Regimes (Mulliner, Harvey, Xia, Fang, and Van Hemert)
Identifying economic regimes in real-time is challenging. This paper presents a systematic approach based on economic variables that compares current market conditions to similar historical periods to guide factor timing. Analyzing past performance after these similar regimes, it determines whether to take long or short positions in equity factors.
Dissecting the return-predicting power of risk-neutral variance (Lu and Pyun)
Investors often look for reliable ways to predict stock returns, but a commonly used measure, risk-neutral variance, may not be as effective as previously thought. The paper finds that past research overstated its predictive power due to look-ahead biases. While it doesn't consistently outperform other models, it becomes more useful during volatile markets. For investors, this suggests that risk-neutral variance is most relevant when market uncertainty is high.
Twitter Follower Growth: Social Media Engagement as Investment Indicators (Pyun)
Investors increasingly turn to social media for market insights, and follower growth on Twitter (now X) appears to signal stock performance. Firms with rapid follower increases, especially in consumer and tech sectors, tend to outperform. A long-short strategy that goes long firms with high follower growth and shorts a market benchmark delivers superior returns, particularly with frequent rebalancing. Tracking social media trends may help investors spot rising stocks before prices fully reflect market attention.
Predicting Market Returns with Book-to-Market Ratios from Institutional trades (Mihai)
Numerous papers have examined the ability of the aggregate book-to-market ratio to forecast market returns. This paper refines the ratio by computing an aggregate book-to-market measure based only on stocks with the highest proportion of institutional buying, better capturing future cash flow expectations. By focusing on this subset of stocks, investors can improve their ability to predict market returns.
ETFs
Hidden Cost of ETF Investing: Retail Demand Shocks and Limits to Arbitrage (Liu, Zhang, and Zhang)
Across a broad range of ETFs, overnight returns are consistently positive, while intraday returns tend to be negative. This pattern isn’t explained by risk or news but arises from retail investors pushing up prices at the open, while arbitrage constraints delay correction.
Multi-day Return Properties of Leveraged Index ETFs (Wang)
The paper studies whether leveraged ETFs (LETFs), which reset daily, suffer from long-term value erosion compared to similar portfolios without daily rebalancing. Contrary to common criticism, the study finds that LETFs generally track their intended leverage over multi-day periods, with deviations occurring mainly in high-volatility environments. These deviations tend to be positively skewed, meaning LETFs are more likely to outperform rather than underperform. The author concludes that investors can use LETFs for leveraged exposure or hedging over periods longer than a day without excessive risk of underperformance.
Machine Learning and Large Language Models
From Econometrics to Machine Learning: Transforming Empirical Asset Pricing (Shi)
This paper provides a broad overview of how machine learning is transforming empirical asset pricing, moving beyond traditional factor models that struggle with complex relationships and high-dimensional data. While classic models prioritize interpretability, machine learning can improve predictive power but raises concerns about economic meaning.
Large language models in finance : what is financial sentiment? (Kirtac and Germano)
Market sentiment influences stock prices, but traditional methods often struggle with nuanced financial text. This review paper explores how large language models improve sentiment analysis, comparing BERT-based models for structured classification and GPT-style models for real-time interpretation. By summarizing recent advancements, it highlights how AI-driven sentiment insights can enhance stock return forecasts and quantitative trading strategies.
Advanced Deep Learning Techniques for Analyzing Earnings Call Transcripts: Methodologies and Applications (Abowath, Daykin, Diagne, and Faile)
Earnings calls provide valuable insights into company performance, but executives often use carefully crafted language that makes sentiment analysis challenging. This paper compares deep learning models, BERT, FinBERT, ULMFiT, and Longformer, to see if they can extract meaningful signals from transcripts and predict stock movements. While FinBERT performs best, the models struggle with overly optimistic corporate language.
Market Behavior
Trends and Reversion in Financial Markets on Time Scales from Minutes to Decades (Safari and Schmidhuber)
Markets alternate between trending and mean-reverting phases, but the pattern varies by time horizon and asset class. Analyzing centuries of financial data, the paper finds that trends persist from hours to years, while shorter and longer time frames often exhibit mean reversion.
Volatility
Forecasting Realized Volatility with Implied Volatility Surface: An Image-Based Approach (Jinting, Xia, and Wuyi)
Many previous studies have converted stock price data into images and used convolutional neural networks (CNNs) to predict returns. This paper takes a different approach by transforming the implied volatility surface into an image-like format and applying CNNs to forecast realized volatility. The model significantly improves volatility predictions compared to traditional methods.
Blogs
Can Margin Debt Help Predict SPY’s Growth & Bear Markets? (Quantpedia)
Volatility Forecasting: HExp Model (Portfolio Optimizer)
Short-Term Mean Reversion in Equity ETFs (QuantSeeker)
GitHub
FinanceDatabase A database of 300.000+ symbols containing Equities, ETFs, Funds, Indices, Currencies, Cryptocurrencies, and Money Markets.
Algorithmic Trading with Python - A course in algo trading.
EliteQuant - A list of online resources for quantitative modeling, trading, and portfolio management.
Medium
10 MindBlowing Free APIs to Supercharge Your Next Project (Parashar)
Implied Volatility Analysis for Insights on Market Sentiment (Velasquez)
18 Insanely Useful Python Automation Scripts I Use Everyday (Parashar)
Podcasts
The Case for Index Funds (Rational Reminder)
Seven Great Investors Break Down the Struggles of Value Investing (Excess Returns)
Owen Lamont, Senior Vice President and Portfolio Manager, Acadian Asset Management (Alpha Exchange)
Last Week’s Most Popular Links
Very.... slow... mean reversion .... and some thoughts on trading at different speeds (Rob Carver)
Volatility of Price-Earnings Ratio and Return Predictability (Jiang and Li)
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.