New Strategies:
#1252 – Systematic Trend-Following with Adaptive Portfolio Construction in Cryptocurrencies
Period of rebalancing: Intraday
Markets traded: cryptos
Instruments used for trading: cryptos
Complexity: Very Complex
Backtest period: 2022-2024
Indicative performance: 40.5%
Estimated volatility: 16.8%
Source paper:
Bui, Duc and Nguyen, Thanh: Systematic Trend-Following with Adaptive Portfolio Construction: Enhancing Risk-Adjusted Alpha in Cryptocurrency Markets
https://arxiv.org/abs/2602.11708
Abstract: Cryptocurrency markets exhibit pronounced momentum effects and regime-dependent volatility, presenting both opportunities and challenges for systematic trading strategies. We propose AdaptiveTrend, a multi-component algorithmic trading framework that integrates high-frequency trend-following on 6-hour intervals with monthly adaptive portfolio construction and asymmetric long-short capital allocation. Our framework introduces three key innovations: (1) a dynamic trailing stop mechanism calibrated to intra-day volatility regimes, (2) a rolling Sharpe-ratio-based asset selection procedure with market-capitalization-aware filtering, and (3) a theoretically motivated asymmetric 70/30 long-short allocation scheme grounded in the empirical positive drift of crypto markets. Through extensive out-of-sample backtesting across 150+ cryptocurrency pairs over a 36-month evaluation window (2022–2024), AdaptiveTrend achieves an annualized Sharpe ratio of 2.41, a maximum drawdown of−12.7%, and a Calmar ratio of 3.18, significantly outperforming benchmark trend-following strategies (TSMOM, time-series momentum) and equal-weighted buy-and-hold portfolios. We further conduct rigorous robustness analyses including parameter sensitivity, transaction cost modeling, and regime-conditional performance decomposition, demonstrating the strategy’s resilience across bull, bear, and sideways market conditions.
#1253 – A Regime-Aware Market Capitalization Rotation Strategy
Period of rebalancing: Monthly
Markets traded: equities, bonds
Instruments used for trading: ETFs
Complexity: Moderate
Backtest period: 2005-2025
Indicative performance: 12.08%
Estimated volatility: 12.77%
Source paper:
Richman, Peter B.: A Regime-Aware Market Capitalization Rotation Strategy Using Hybrid Asset Allocation Momentum
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6096427
Abstract: This paper examines a regime-aware market-capitalization rotation strategy implemented within the Hybrid Asset Allocation (HAA) framework. The strategy allocates among large-, mid-, and small-cap U.S. equity segments (SPY, MDY, IJR) using exchange-traded funds and dynamically adjusts exposure based on market regime conditions. Regime determination is governed by the absolute momentum of an inflation-protected bond ETF (TIP), defined as a multi-horizon composite measure calculated as the unweighted average of the trailing 1-, 3-, 6-, and 12-month total returns (13612). When this regime signal is non-positive, the portfolio shifts entirely to short-duration Treasury bills (SHY). During favorable regimes, equity selection is based on the same 13612 momentum measure to target the best-performing market-cap segment. Performance is evaluated relative to a buy-and-hold investment in the total U.S. equity market (VTI) over a 21-year sample period from January 2005 to December 2025. The results indicate that the regime-aware market-cap rotation strategy achieves higher compound returns (12.08% gross; 10.88% net), lower volatility (12.77%), and substantially reduced maximum drawdowns (-20.01% vs. -50.84% for the benchmark). This study demonstrates that applying a parsimonious HAA-style framework to intra-equity market segmentation can materially improve downside risk management and risk-adjusted performance (Sharpe ratio of 0.84 net vs. 0.74) while maintaining competitive long-term growth.
#1254 – Government Shutdown Signal Strategy
Period of rebalancing: Daily
Markets traded: equities
Instruments used for trading: ETFs, funds, futures, CFDs
Complexity: Moderate
Backtest period: 1980-2025
Indicative performance: 1.2%
Estimated volatility: –
Source paper:
Belobrad, David and Vojtko, Radovan: Can We Use U.S. Government Shutdowns as a Signal for Investment Decisions?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6196863
Abstract: In recent times, we have observed heightened volatility across financial markets. Concerns surrounding government shutdowns, as well as the uncertainty they create, do little to calm these fluctuations. Rather than being purely disruptive, however, such events raise an intriguing question: could these episodes of political and economic uncertainty actually be leveraged to our advantage in investment strategies? In this article, we will examine several asset classes and attempt to assess whether this phenomenon provides a sufficiently relevant signal for investment decisions.
#1255 – Machine Learning Insider Trading Strategy
Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Very Complex
Backtest period: 2002-2023
Indicative performance: 10.03%
Estimated volatility: 18.4%
Source paper:
Kong, Tongyang and Wei, Chishen and Zhang, Lei: Can Machines Learn from Trading Data?
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5967134
Abstract: We horse race the ability of human intelligence versus machine learning algorithms in extracting trading signals from raw trading data. Using publicly available corporate insider trades from SEC Form 4 filings, we train a gradient-boosted decision tree model to predict future stock returns. The long-short portfolio has an annual Sharpe ratio of 1.09 and earns a monthly return of 1.14%, which is not explained by standard risk factors. Our machine-generated signal provides significant incremental information distinct from existing human-generated insider trading signals and exhibits stronger predictive ability in hard-to-value stocks. Our evidence indicates that machines discover information embedded in insider purchases beyond human intelligence, while humans uncover information in insider sales that machines cannot detect.
#1256 – Agentic AI Nowcasting Predicts Stock Returns
Period of rebalancing: Daily
Markets traded: equities
Instruments used for trading: stocks
Complexity: Very Complex
Backtest period: 2025-2026
Indicative performance: 57.46%
Estimated volatility: 72.4%
Source paper:
Chen, Zefeng and Pu, Darcy: Autonomous Market Intelligence: Agentic AI Nowcasting Predicts Stock Returns
https://arxiv.org/abs/2601.11958
Abstract: Can fully agentic AI nowcast stock returns? We deploy a state-of-the-art Large Language Model to evaluate the attractiveness of each Russell 1000 stock daily, starting from April 2025 when AI web interfaces enabled real-time search. Our data contribution is unique along three dimensions. First, the nowcasting framework is completely out-of-sample and free of look-ahead bias by construction: predictions are collected at the current edge of time, ensuring the AI has no knowledge of future outcomes. Second, this temporal design is irreproducible—once the information environment passes, it can never be recreated. Third, our framework is 100% agentic: we do not feed the model news, disclosures, or curated text; it autonomously searches the web, filters sources, and synthesises information into quantitative predictions. We find that AI possesses genuine stock selection ability, but only for identifying top winners. Longing the 20 highest-ranked stocks generates a daily Fama-French five-factor plus momentum alpha of 18.4 basis points and an annualised Sharpe ratio of 2.43. Critically, these returns derive from an implementable strategy trading highly liquid Russell 1000 constituents, with transaction costs representing less than 10% of gross alpha. However, this predictability is highly concentrated: expanding beyond the top tier rapidly dilutes alpha, and bottom-ranked stocks exhibit returns statistically indistinguishable from the market. We hypothesise that this asymmetry reflects online information structure: genuinely positive news generates coherent signals, while negative news is contaminated by strategic corporate obfuscation and social media noise.
#1257 – Atypical Firms and the Cross-Section of Expected Stock Returns
Period of rebalancing: Monthly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Very Complex
Backtest period: 1981-2023
Indicative performance: 19.99%
Estimated volatility: 49.79%
Source paper:
Weibels, Sebastian: Hard to Process: Atypical Firms and the Cross-Section of Expected Stock Returns
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6048715
Abstract: Theories of limited attention predict that investors rely on typical patterns to navigate high-dimensional firm characteristics, making atypical firms hard to process. To quantify this difficulty, we propose a data-driven measure of firm atypicality using an autoencoder (ATYP). The model learns typical patterns that describe most firms, and our measure aggregates the deviations those patterns cannot explain. Unlike proxies based on disclosure or organizational complexity, this approach captures the processing difficulty of the characteristics themselves. Empirically, we document that atypicality strongly predicts future returns. A decile portfolio that sells high-ATYP firms and buys low-ATYP firms earns 1.47% per month (equal-weighted) and 0.82% (value-weighted). The effect strengthens where investor attention is low and arbitrage is limited, suggesting mispricing as the explanation.
#1258 – Morning Volatility Uncertainty and Variance Risk Premium
Period of rebalancing: Intraday
Markets traded: equities
Instruments used for trading: futures, options
Complexity: Very Complex
Backtest period: 2013-2022
Indicative performance: 16.07%
Estimated volatility: 10%
Source paper:
Bevilacquaa, Mattia and Hizmeria, Rodrigo: Early Birds Get the Vol: Morning Volatility Uncertainty and Variance Risk Premium
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6212458
Abstract: We document that morning volatility-of-volatility (VVIX), measured at 10:00 EST during the U.S.–European market overlap, strongly predicts next-day variance asset returns. Predictive power peaks with t-statistics reaching 5.6 and adjusted R2 of 2.6%, but diminishes after 11:00 EST. Trading strategies exploiting this pattern generate Sharpe ratios exceeding 2.0, remaining economically meaningful even after transaction costs. Crucially, intraday returns display a flat relationship with morning VVIX on high volatility uncertainty days, while close-to-close returns exhibit strong predictability. These findings are difficult to reconcile with standard risk premium theories but are consistent with limited attention and slow-moving beliefs about volatility.
New research papers related to existing strategies:
#961 – Salient Theory Predicts US Stocks in the Cross Section
Guo, Jiaqi and Li, Youwei: Salience Theory and Risk Anomalies
https://ssrn.com/abstract=4603171
Abstract: Investor demand for stocks is driven by their salient returns (Bordalo, Gennaioli, and Shleifer, 2012). We hypothesize that high-risk stocks are more likely to yield salient returns than low-risk stocks. Using various risk measures, we show that the risk-return relation is significantly positive for stocks with salient downsides but negative for stocks with salient upsides. We provide evidence that retail investors pay strong attention to salient stock payoffs and demand more (less) high-risk stocks in the presence of salient upsides (downsides). Our findings hold after controlling for alternative explanations including lottery preference, arbitrage risk, reference-dependent preference, disagreement and coskewness risk.
#826 – News Sentiment and Equity Returns – BERT ML Model
Mishra, Suketu and Girase, Rakshit Girase and Patil, Sarthak Patil and Chaudhari, Vedant and Jadhav, Supriya: Stock Predictor: Predicting Stock Movements Based on Current Global News Using Artificial Intelligence
https://ssrn.com/abstract=5727363
Abstract: The prediction of stock price dynamics, a central challenge in financial econometrics, is increasingly influenced by the continuous flow of information from global news that shapes investor sentiment. This research quantifies the incremental predictive power of news by proposing a multimodal deep learning framework that systematically fuses sentiment signals, derived from real-time global news, with historical market data to forecast short-term stock movements. The proposed architecture leverages a domainspecific language model, FinBERT, for robust textual feature extraction, which is then integrated with advanced sequence models, including LSTMs and transformers, to model the temporal interplay between market and sentiment dynamics. Through rigorous empirical evaluation on a diverse set of stock indices and individual equities from 2021-2024, our findings demonstrate that these sentiment-augmented models achieve a significant performance gain over traditional time-series baselines. Specifically, our topperforming transformer-based fusion model registers a directional accuracy of 63.2%, a notable improvement over the 56.5% accuracy of a baseline LSTM relying solely on price data. The analysis confirms that news sentiment is a particularly potent predictive feature during periods of high market volatility. This work concludes with a critical discussion on inherent limitations, such as signal noise and causality challenges, while outlining future research directions for enhancing model explainability and real-world deployment readiness.
#563 – Currency Factor Momentum
Bartel, Merlin and Hanke, Michael and Petric, Sebastian: FX Factor Momentum Pre- and Post-GFC
https://ssrn.com/abstract=6432743
Abstract: Building on the well-documented role of the dollar and carry factors in global currency research, this paper investigates FX factor momentum in a systematic manner. We analyze the performance of factor momentum strategies on various sets of developed and emerging market currencies. We find that the performance of FX factor momentum strategies not only depends on the time period (pre- vs. post-GFC), but also on the particular set of currencies used to construct the factors. Comparing statistical factors from PCA across time reveals changes in the cross-section of currencies that affect the performance of these strategies.
And several interesting free blog posts that have been published during the last month:
Anomaly-Based Trading Strategies in the Real Estate Sector. Can the Market Be Beaten?
This study examines the effectiveness of several anomaly-based trading strategies applied to the real estate sector represented by the RlEst index from the Fama–French 48 industry portfolios. Using monthly data from July 1, 1926, to December 1, 2025, we analyze whether selected strategies are capable of generating superior risk-adjusted returns compared to both the standalone RlEst index and the broader market represented by the Fama–French 12-industry portfolios. The tested approaches include trend-following strategies based on moving averages, momentum strategies based on the rate of change of the index, and seasonality-based strategies utilizing different look-back periods.
Timing Value vs. Growth: Evidence from 100 Years of Small Value–Large Growth Spread
The goal of our article is to examine the long-term relationship between small value and large growth stocks using more than 100 years of data and test whether the spread between small value and large growth portfolios shows trends that could help investors switch between the two styles. Using the Fama and French 2×3 and 5×5 size and book-to-market portfolios, we construct the small value minus large growth (SV–LG) spread and apply simple trend-following signals based on moving averages and momentum with horizons ranging from 3 to 12 months. Our results show that trend-following strategies are able to capture part of the value outperformance on the long side. Timing periods when growth stocks dominate is much more difficult.
Quantpedia’s Research Workflow: From Idea Discovery to Portfolio Construction
Quantitative strategy research is rarely about discovering a single “perfect” trading rule. In practice, robust portfolios emerge from a structured research process that filters ideas, evaluates evidence, and combines complementary strategies.
In this article, we demonstrate how such a workflow can be implemented using the tools available in Quantpedia Pro. Rather than focusing on maximizing the performance of a single strategy, we walk through the research process step by step—from thematic filtering to portfolio-level evaluation.
To make the process concrete, we use value-based equity strategies as our working example. However, the goal of the article is not to identify the ultimate value strategy, but to illustrate how a systematic research workflow can be used to build a diversified portfolio of strategies around any investment hypothesis.
The Illusion of the Carbon Premium
Carbon that has not yet been emitted should not be used to predict stock returns. While this sounds obvious, prior research papers have done exactly that. This critical observation forms the basis for the Robeco Institutional Asset Management research team’s re-examination of the relationship between climate risk and asset pricing. Investors and academics alike have sought to understand how environmental factors influence stock returns, often assuming that higher emitters command a risk premium. However, the timing of data availability is crucial in quantitative strategy formation, and misalignments here can lead to spurious conclusions about the pricing of carbon emissions.
Plus, the following trading strategies have been backtested in QuantConnect in the previous two weeks:
1244 – Systematic FX Session Breakout Strategy
1247 – Momentum-Driven EAFE Tactical Allocation
1248 – FOMC Cycle 2-year Treasury Momentum Strategy
