Quantpedia Premium Update – May 4th

New Strategies:

#995 – Pre-Refunding Announcement Gains in U.S. Treasuries

Period of rebalancing: Daily
Markets traded: bonds
Instruments used for trading: bonds, CFDs, ETFs, forwards, futures, swaps
Complexity: Simple strategy
Backtest period: 1991-2023
Indicative performance: 0.94%
Estimated volatility: 1.7%

Source paper:

Wang, Chen and Zhao, Kevin, Pre-Refunding Announcement Gains in U.S. Treasurys
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4764295
Abstract:
Each quarter, the Treasury Department unveils its refunding plan, detailing the following quarter’s treasury issuances in terms of size and maturity composition. We document substantial positive returns on long-term Treasurys on the day before these Treasury Refunding Announcements (TRAs), a pattern persisting since the 1990s and intensifying over the last two decades amidst growing Federal deficits. These pre-TRA gains are distinct from known end-of-month pricing patterns and account for a sizable fraction of annual yield and term premium changes. Implementing a trading strategy focused solely on these four days per year yields a Sharpe ratio of over 4. We provide evidence of uncertainty reduction and associated information production around TRAs as a potential mechanism. Finally, we discuss implications for some documented bond market patterns and the pre-FOMC drift in the equities market.

#996 – Macroeconomic Momentum in the Cross-Sectional Equity Market Indices

Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: CFDs, ETFs, funds, futures
Complexity: Complex strategy
Backtest period: 1995-2020
Indicative performance: 3.6%
Estimated volatility: 4.14%

Source paper:

Yu Zhang, Konstantina Kappou, Andrew Urquhart: Macroeconomic momentum and cross-sectional equity market indices
https://www.sciencedirect.com/science/article/pii/S1042443124000404
Abstract:
Momentum is a well-known and studied artefact of financial markets. In this paper, we investigate whether momentum in a country’s macroeconomic variables is related to the future performance of equities in that country. We find that the past economic trends of a country’s fundamentals are positively associated with the equity market index returns. Based on that, an economic momentum portfolio of buying (selling) equity index in countries with relatively strong (weak) economic past trends exhibits an annualised Sharpe ratio of 0.87. The economic momentum portfolio outperforms benchmarks regarding rewards to variability and maximum drawdown and yields an annualised alpha of 3.72%, leaving 95% of the returns unexplained by the benchmarks.

#997 – Clustering Based Multi-Pairs Trading

Period of rebalancing: Daily
Markets traded: equities
Instruments used for trading: stocks
Complexity: Very complex strategy
Backtest period: 2000-2022
Indicative performance: 10.99%
Estimated volatility: 10.77%

Source paper:

Cartea, A., and Cucuringu, and M. and Jin, Q..: Correlation Matrix Clustering for Statistical Arbitrage Portfolios
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4560455
Abstract:
We propose a framework to construct statistical arbitrage portfolios with graph clustering algorithms. First, we use various clustering methods to partition the correlation matrix of market residual returns of stocks into clusters. Next, we construct and evaluate the performance of mean-reverting statistical arbitrage portfolios within each cluster. We explore five clustering algorithms and demonstrate that our proposed framework generates profitable trading strategies with over 10% annualized returns and statistically significant Sharpe ratios above one. The performance of our statistical arbitrage portfolios is neutral to the market and cannot be fully explained by intra-industry mean-reversion effects.

#998 – Google Trends Sentiment as a Predictor for Cryptocurrency Returns

Period of rebalancing: Monthly
Markets traded: cryptos
Instruments used for trading: CFDs, cryptos, ETFs, funds, futures
Complexity: Simple strategy
Backtest period: 2017-2023
Indicative performance: 74.58%
Estimated volatility: 61.98%

Source paper:

Zelieska, Lukáš and Vojtko, Radovan and Dujava, Cyril: Can Google Trends Sentiment Be Useful as a Predictor for Cryptocurrency Returns?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4806394
Abstract:
In the fast-paced world of cryptocurrencies, understanding market sentiment can provide a crucial edge. As investors and traders seek to anticipate the volatile movements of Bitcoin, innovative approaches are continuously explored. One such method involves leveraging Google Trends data to gauge public interest and sentiment towards Bitcoin. This approach assumes that search volume on Google not only reflects current interest but can also serve as a predictive tool for future price movements. This blog post delves into the intricacies of using Google Trends as a sentiment predictor, exploring its potential to forecast Bitcoin prices and discussing the broader implications of sentiment analysis in the financial market.

#999 – Bottom-Up LASSO Strategy Using 10K Reports to Predict Stocks’ Performance

Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Very complex strategy
Backtest period: 1995 -2021
Indicative performance: 5.91%
Estimated volatility: 8.1%

Source paper:

Ross, L., Horn, J., Pilanci, M., Luo, K. and Zhou, G..: What Does Firm 10-K Tell Us?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4493166
Abstract:
In contrast to the recent increasing focus on large languages model, we propose a bottom-up approach that exploits the individual predictive power of each word. Our word dictionary is constructed by using a data-driven approach, and it is these selected words that are used to build the predictive model with lasso regularized regressions and large panels of word counts. We find that our approach effectively estimates the cross-section of stocks’ expected returns, so that a factor that summarizes the information generates economically and statistically significant returns, and these returns are largely unexplained by standard factor models. However, an inspection of the factor dictionary indicates the element contains many words with possible risk-related interpretations, such as currency, oil, research, and restructuring, which increase a stock’s expected return, while the words acquisition, completed, derivatives, and quality decrease the expected return.

#1000 – Sentiment trading with large language models

Period of rebalancing: Daily
Markets traded: equities
Instruments used for trading: stocks
Complexity: Very complex strategy
Backtest period: 2010 -2023
Indicative performance: 137.5%
Estimated volatility: 45%

Kirtac, K., and Germano, G.: Sentiment trading with large language models
https://www.sciencedirect.com/science/article/pii/S1544612324002575?ssrnid=4706629&dgcid=SSRN_redirect_SD
Abstract:
We analyse the performance of the large language models (LLMs) OPT, BERT, and FinBERT, alongside the traditional Loughran-McDonald dictionary, in the sentiment analysis of 965,375 U.S. financial news articles from 2010 to 2023. Our findings reveal that the GPT-3-based OPT model significantly outperforms the others, predicting stock market returns with an accuracy of 74.4%. A long-short strategy based on OPT, accounting for 10 basis points (bps) in transaction costs, yields an exceptional Sharpe ratio of 3.05. From August 2021 to July 2023, this strategy produces an impressive 355% gain, outperforming other strategies and traditional market portfolios. This underscores the transformative potential of LLMs in financial market prediction and portfolio management and the necessity of employing sophisticated language models to develop effective investment strategies based on news sentiment.

#1001 – AI Predicts Stock Returns with Company Filings

Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Very complex strategy
Backtest period: 2003 -2022
Indicative performance: 10.24%
Estimated volatility: 10.24%

Source paper:

Chapados, N., and Fan, Z., and Goyenko, R., and Laradji, and I.H., Liu, and F. and Zhang, C.: Can AI Read the Minds of Corporate Executives?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4493166
Abstract:
It can. Using textual information from a complete history of regular quarterly and annual filings by U.S. corporations, we train classic machine learning algorithms and large language models, LLMs, to predict future earnings surprises. We first find that the length of MD&A section on its own is negatively associated with future earnings surprises and firm returns in the cross-section. Second, neither sentiment-based nor bag-of-words classic machine learning regression-based approaches are able to “learn” from the past managerial discussions to forecast future earnings. Third, only finance-objective trained LLMs have the capacity to “understand” the contexts of previous 10-Q (10-K) releases to predict both positive and negative earnings surprises, and subsequent future firm returns. We find significant, and often hidden in the complexity of presentations, positive and negative informational content of publicly disclosed corporate filings, and superior (to human and classic NLP approaches) abilities of more recent AI models to identify it.

New research papers related to existing strategies:

#971 – International Market Timing With Moving Average Distance

Li, Yuanpeng and Luo, Yan and Xiao, Zhiguo: Moving Average as a Psychological Barrier: Evidence from International Markets
https://ssrn.com/abstract=4486459
Abstract:
We show that the moving averages (MAs) of stock market indices act as psychological barriers. We reveal that market indices do not move continuously near their MAs, especially in markets dominated by unsophisticated investors. Using international data, we find that a stock market index’s MA exerts a significant impact on future index returns when it is crossed over, and the effect is distinct from both the 52-week high effect of George and Hwang (2004) and the historical high effect of Li and Yu (2012). The MA effect is stronger when investors’ sophistication is lower or when their anchoring propensity is higher. A trading strategy formulated based on the MA effect generates an average abnormal annual return of 1.69% to 4.26% in international markets.

#382 – The 52-Week High and Short-Term Reversal in Stock Returns

Kim, Somyung and Ohk, Kiyool: The Influence of Psychological Price Barriers on Short-Term Return Reversals
https://ssrn.com/abstract=4772185
Abstract:
This study explores short-term return reversals in stock markets, this phenomenon that has not been adequately explained in the finance literature. The study highlights the role of psychological price barriers in the understanding of short-term return reversals. The anchoring effect, in which investors fixate on a reference point, such as the 52-week high, plays a crucial role. We use nearness past 52 weeks high (NH52) as a proxy for psychological price barriers. Proximity to this psychological barrier influences investor reactions; nearness prompts pessimism, undervaluing positive news, while distance induces optimism, leading to overvaluation. The findings suggests that short-term reversal strategies are more profitable for stocks distant from their 52-week highs.Contrary to the previous emphasis on liquidity, this study links short-term reversals to psychological factors—specifically, over- and under-reactions driven by psychological price barriers. Low-NH52 stocks exhibit stronger short-term reversals, which are attributed to overreactions to both good and bad news. The interplay between overreactions to positive news and overreactions to liquidity risk results in pronounced short-term reversals. This study utilizes double-sorting portfolio strategies based on winners and losers and confirms that short-term reversals are more prominent in low-NH52 stocks. This finding supports the significance of psychological price barriers, liquidity supply, and investor risk aversion in understanding short-term return reversals. This study underscores the potential impact of psychological factors, along with liquidity considerations, in explaining market anomalies.

#42 – Alpha Cloning – Following 13F Fillings

Schroeder, J and Posch, Peter N.: Outperforming the Market: Portfolio Strategy Cloning from SEC 13F Filings
https://ssrn.com/abstract=4767576
Abstract:
Can mirroring the investment strategies of institutional managers lead to market-outperforming returns? Our findings demonstrate that cloned portfolios in the top quartile, derived from SEC EDGAR Form 13F filings, replicate the funds’ performances and exceed the SP500 index by 24.25% on an annualized risk-adjusted basis. Analyzing over 150,000 portfolios between 2013 and 2023, we compare original versus replicated strategies across 12 metrics, such as Alpha, Sharpe and Sortino ratios, various return rates, annualized volatility, and maximum drawdown. Through Wilcoxon signed-rank tests applied to delta distributions, we reject the null hypothesis across all metrics, except for annualized volatility, maximum drawdown and tracking error, and demonstrate that cloned portfolios balanced on the disclosure date of filings (rather than quarter-end), successfully mirror the performance of the original funds, including both market-underperforming and -outperforming funds.

#605 – Momentum on Straddles

Heston, Steven L. and Jones, Christopher S. and Khorram, Mehdi and Li, Shuaiqi and Mo, Haitao: Option Momentum
https://ssrn.com/abstract=4113680
Abstract:
This paper investigates the performance of option investments across different stocks by computing monthly returns on at-the-money straddles on individual equities. It finds that options with high historical returns continue to significantly outperform options with low historical returns over horizons ranging from 6 to 36 months. This phenomenon is robust to including out-of-the-money options or delta-hedging the returns. Unlike stock momentum, option return continuation is not followed by long-run reversal. Significant returns remain after factor risk adjustment and after controlling for implied volatility and other characteristics. Across stocks, trading costs are unrelated to the magnitude of momentum profits.

#498 – Value in Anomalies

Zhang, Shaojun: Understanding Factor Value
https://ssrn.com/abstract=4759616
Abstract:
The value spread of factors fluctuates over time because of changes in market equity or book value but predicts factor returns only through the component driven by market equity changes (the dme spread). Exploiting cross-sectional mispricing, the dme spread captures 90 years of sentiment and subsumes the predictability in existing sentiment measure. Factor predictability concentrates on factors most predictable by sentiment and factors more subject to asymmetric limits of arbitrage. A factor value strategy exploiting the predictability outperforms and explains cross-sectional value factors. The value premium is not an independent factor but summarizes time-varying factor returns conditional on sentiment.

#65 – Enhanced Value Premium
#376 – Combining Fundamental FSCORE and Equity Short-Term Reversals

Hasan, Iftekhar and Shen, Jianfu and Ng, Chi Cheong: Do Institutional Investors Exploit Expectation Errors in Value/Glamour Stocks?
https://ssrn.com/abstract=4761222
Abstract:
This study examines the institutional demand for mispriced stocks with incongruent expectations implied by book-to-market ratio and financial strength. Consistent with the argument of expectation errors in value/glamour stocks (Piotroski and So, 2012), institutional investors buy value stocks with strong fundamentals (underpriced) and sell glamour stocks with weak fundamentals (overpriced). Independent institutions are more likely to take advantage of the mispricing in value/glamour firms than passive institutions. Changes in institutional ownership concentrate on stocks with more limits or restrictions on arbitrage. Institutional trading on expectation errors attenuates the book-to-market anomaly and the abnormal returns to mispriced stocks. Institutional trading patterns on mispriced value/glamour stocks are also documented in global markets.

And several interesting free blog posts that have been published during the last 2 weeks:

ESG Investing during Calm and Crisis Periods

Over the last decade, investing responsibly and deploying capital for “ethically” correct and sustainable growth has been quite a theme. We dedicated a few blogs to this theme and have a separate ESG category for trading strategies in our database. It is often easy to commit financial resources to noble ideas during liquidity abundance. However, how do these methodologies fare during crisis times, such as when the GFC (Global Financial Crisis) or COVID-19 hit? That’s the question that a new paper by Henk Berkman and Mihir Tirodkar tries to answer.

Private vs. Public Investment Strategies

Choosing the right investment strategy plays a crucial in portfolio allocation decisions, particularly when considering both private and public asset classes. While the reported performance of public assets typically matches their real-world performance, the same cannot be said for private assets due to the complexities of fund selection, commitment pacing, and return on uncalled and uncommitted capital. Fortunately, there are ways to incorporate public and private asset classes into one portfolio optimally. One example is the recent paper written by Xiang Xu, which introduces the Fair Comparison (FC) framework, which provides a methodology to measure the real-world performance of private investment strategies.