New Strategies:
#1230 – Intraday First Hour Reversal
Period of rebalancing: Intraday
Markets traded: equities
Instruments used for trading: ETFs
Complexity: Moderate
Backtest period: 1996-2024
Indicative performance: 7.76%
Estimated volatility: 5.13%
Source paper:
Iwanaga, Yasuhiro and Sakemoto, Ryuta: Intraday time series reversal
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5807282
Abstract: This study investigates intraday time series reversal that the overnight return predicts the first half-hour return for U.S. market indices. Intraday reversal strengthens during periods of high volatility, including the global financial crisis and the early phase of the COVID-19 pandemic. In contrast, it weakened after the 2010s and during the later phase of the pandemic. These findings suggest that intraday reversal is associated with the presence of opposing investor clienteles—overnight buying by retail investors and daytime selling by arbitrageurs. Moreover, we do not observe intraday reversal for other ETFs that track international market indices. Finally, we propose investment strategies based on intraday reversal and present that they are profitable for U.S market indices but not for international market indices.
#1231 – Magic Formula in India
Period of rebalancing: Quarterly
Markets traded: equities
Instruments used for trading: stocks
Complexity: Complex
Backtest period: 2007-2024
Indicative performance: 14.19%
Estimated volatility: –
Source paper:
Kandukuri, Vishwesh and Jain, Kashish and Anand, Pratik: Beyond the Benchmark: Magic Formula Outperformance in Indian Equity Markets
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5493626
Abstract: Most existing studies on the Magic Formula strategy have been conducted in the context of developed markets, while research in the Indian setting has remained limited and largely exploratory. The present study implements and evaluates the Magic Formula in the Indian equity market, focusing on NIFTY 100 constituents from 2007 to 2024. The Magic Formula, introduced by Joel Greenblatt, ranks firms based on earnings yield and return on capital to identify undervalued yet profitable companies. Using annual financial data and adjusted stock prices, we construct portfolios following the standard Magic Formula rules and compare their performance with the NIFTY 50 benchmark. Our findings reveal that even the unmodified version of the Magic Formula outperforms the benchmark substantially, generating approximately 1.7 times higher returns over the study period. These results are particularly notable given that the test universe consists of large, liquid, and relatively low-risk firms. The evidence suggests that extending the strategy to broader indices such as the NIFTY 250 or NIFTY 500 could yield even stronger results, underscoring the potential of systematic value-based approaches in Indian equity markets.
#1232 – Defensive Absolute Return Portfolio
Period of rebalancing: Monthly
Markets traded: equities, commodities, bonds, currencies
Instruments used for trading: stocks, futures
Complexity: Complex
Backtest period: 1800-2021
Indicative performance: 2.8%
Estimated volatility: 5%
Source paper:
Baltussen, Guido and Martens, Martin and van der Linden, Lodewijk: The Best Defensive Strategies: Two Centuries of Evidence
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5815464
Abstract: We examine downside protection – or defensive – strategies over more than 220 years of global financial history – covering many years where traditional equity-bond portfolios suffer and across a wide range of economic scenarios and historical regimes. Traditional defensive equity factors – low-risk, quality, and value – consistently provide effective downside protection, whereas gold and put options prove less drawdown- or cost-effective. Our long-run evidence shows that multi-asset defensive strategies, particularly a return-enhanced version of the Defensive Absolute Return (DAR) portfolio introduced by Cavaglia et al. (2022) and trend-following, provide the most effective downside protection. DAR and trend-following are complementary across tests by diversifying each other across stages of drawdowns. Investors can improve the defensive properties and improve total portfolio outcomes of traditional portfolios by considering the deep sample evidence on defensive strategies provided in this paper.
#1233 – Long-Run Interest Rate Differentials and Carry Trade
Period of rebalancing: Monthly
Markets traded: currencies
Instruments used for trading: forwards, futures, CFDs, swaps
Complexity: Moderate
Backtest period: 1987-2023
Indicative performance: 2.48%
Estimated volatility: 6.63%
Source paper:
Kaebi, Mohammed Mehdi and Martins, Igor Ferreira Batista: Long-Run Interest Rate Differentials and the Profitability of Currency Carry
https://ssrn.com/abstract=5910730
Abstract: This paper examines the role of long-run and cyclical components of interest rate differentials in explaining the returns to currency carry strategies. We show that long-run differentials account for most of the profitability, while cyclical differentials play only a limited role. A simple strategy that goes long currencies above the median long-run differential and shorts those below delivers a statistically and economically significant annualized excess return of 2.48%. Relative to traditional carry, our strategy achieves a higher Sharpe ratio, lower turnover and a less negative skewness. We introduce a new tradable carry factor that explains the cross section of currency returns beyond the benchmark carry factor.
A new research paper related to existing strategies:
#118 – Time Series Momentum Effect
Kjaer, Christian: On the Anatomy of Trend
https://ssrn.com/abstract=5957236
Abstract: We present a theoretical framework for analyzing the performance of time-series momentum (trend-following) strategies when the underlying asset follows a general stationary Gaussian process. We derive closed-form expressions for the unconditional expected log payoff, decomposing the performance into a drift component related to the expected position and a timing component related to the alignment of the trend signal and the subsequent return of the underlying. Extending the analysis to conditional performance, we show that the conditional expected payoff preserves the characteristics from the unconditional: a conditional expected position component and a conditional timing component. We derive exact asymptotic limits for the conditional payoff in “deep crash” scenarios over a fixed horizon and show that even with i.i.d. returns, the conditional expected return is positive if the drawdown period is larger than one day. Finally, we show that a sufficiently long and severe drawdown always yields positive trend performance-irrespective of the underlying autocorrelation structure. We also characterize the role of the lookback parameter, revealing a trade-off between expected returns and hedge effectiveness. Empirical analysis of 103 assets over 25 years confirms the key theoretical predictions with rigorous statistical significance. We document a strongly convex conditional payoff profile (“smile”), quantify the “cost of convexity”.
#62 – Shorting Overvalued Stocks
#809 – Value Effect in Unprofitable Firms
Cheng, Zhuo and Fang, Jing and Huang, Shiqi: Financial Distress and Stock Comovement
https://ssrn.com/abstract=5960346
Abstract: Contrary to the financial distress premium hypothesis, stocks of financially distressed firms exhibit the lowest return comovement. These firms are characterized by high valuation uncertainty and information and arbitrage frictions. As a result, their stocks are prone to substantial mispricing and tend to be driven by firm-specific factors rather than systematic (factor-wide) forces—leading to highly idiosyncratic price behavior. Using a linear factor model, we show that this heightened idiosyncratic variation accounts for their low comovement. Moreover, consistent with the positive relation between financial distress and the degree of mispricing (undervaluation and overvaluation), financially distressed firms are more likely to experience extreme positive or negative returns. Recognizing these dynamics helps resolve asset pricing puzzles related to financial distress—such as the high returns generated by some anomaly-based trading strategies when applied to financially distressed firms and the inconsistent empirical findings on the distress-return relation.
#615 – Machine Learning and Mutual Fund Characteristics
Fausch, Jürg and Frigg, Moreno and Johann, Thomas and Mussbach, Emil and Westheide, Christian and Drobetz, Wolfgang: Does machine learning really help to select mutual funds with positive alpha?
https://ssrn.com/abstract=5937794
Abstract: We revisit the recent finding of DeMiguel, Gil-Bazo, Nogales and Santos (2023) that machine learning (ML) methods can identify long-only portfolios of actively managed mutual funds with positive net annual alpha of up to 2.69%. Using their replication package, we show that these conclusions are driven by a coding error that introduces a forward-looking bias in the construction of portfolio returns. Removing the bias eliminates the long-only alpha. Based on an independent replication, we find that nonlinear ML methods consistently identify underperforming funds. However, this also holds for linear models. Consequently, long–short portfolios deliver robust and economically meaningful alphas, driven predominantly by the short leg. We further extend the analysis by expanding the sample period to the end of 2024, by considering forecast horizons up to 36 months and conducting univariate sorts across fund characteristics. These extensions do not materially enhance long-only predictability but offer empirical evidence of the relative advantages of ML at longer horizons. Overall, our results suggest that ML adds value primarily through the consistent avoidance of poorly performing funds rather than by identifying long-only outperforming ones. This has important implications for both the academic interpretation of ML performance and its practical application in active mutual fund selection.
#1083 – Adjusted Momentum Strategies in Indian Stocks
Kumar, Arnav: Cross-Sectional Momentum in Indian Equities: A Reproducible Evaluation on the Nifty-50
https://ssrn.com/abstract=5744965
Abstract: This paper presents a reproducible empirical study of the cross-sectional momentum anomaly within India’s Nifty-50 constituents from 2012-2025. Using monthly adjusted prices, we evaluate both a baseline (equal-weighted, long-short) and an optimized (volatility-scaled, long-only) implementation. The optimized specification achieves an annualized Sharpe ratio above 3 in-sample even after transaction costs, and delivers strong out-of-sample performance (ann. return ≈ 133.6%, vol ≈ 46.1%, Sharpe ≈ 2.90, max drawdown ≈ 41.6%). These results suggest that medium-term momentum in Indian equities arises from under-reaction and sectoral trend persistence, consistent with behavioral and informational frictions documented in emerging markets.
And several interesting free blog posts that have been published during the last month:
The Fallacy of Concentration Risk
Market concentration has become one of the most discussed structural risks in today’s equity markets. A small group of mega-cap stocks—often the largest five to ten names—now accounts for an unusually large share of major market indices. This has led to widespread concerns that such concentration makes markets more fragile and that elevated index weights at the top may foreshadow weaker future returns. Many investors worry that history is repeating itself and that extreme concentration today implies disappointment tomorrow.
Is The Optimal Long-term Portfolio Share of Bitcoin Negative?
The crypto-enthusiast’s mantra—“just add Bitcoin and watch the efficient frontier fly”—runs into a hard empirical wall when you extend the sample, tighten the econometrics, and force the asset to compete on identical risk-adjusted footing with equities. Alistair Milne’s new SSRN paper applies a textbook Markowitz mean–variance framework to a two-asset universe (S&P 500 vs. Bitcoin) and finds that the ex-ante optimal long-term weight on BTC is not merely small; it is outright negative. In other words, a rational, variance-averse allocator who believes expected returns equal historical equity premia plus a fair compensation for BTC’s non-diversifiable volatility should be short, not long, the flagship digital token.
Who Is the Counterparty to the Pro-Cyclical Investors
An interesting transaction-level study we take a closer look at today asks who takes the other side of trades when the most pro-cyclical players in markets — primarily asset managers — buy in booms and sell in busts. The paper uses comprehensive transaction data across major European equity and interest-rate cash and derivatives markets to classify counterparties by sector and to measure, at horizons from 15 minutes to one month, which sectors absorb net flows from pro-cyclical investors. Dealer banks emerge as the dominant liquidity providers across asset classes. At intraday and daily horizons, dealer banks absorb the vast majority of the net flow coming from asset managers. Other active liquidity sources, such as principal trading firms and hedge funds, play only minor roles.
Plus, the following trading strategies have been backtested in QuantConnect in the previous two weeks:
1200 – Using Short interest on Leveraged ETFs for Timing the Stock Index Returns
1201 – One-Month Factor Momentum Strategy
1210 – The End-Of-Month Effect in Value-Growth and Real-Estate-Equity Spreads
1222 – Box Index Strategy
1228 – Multi-Timeframe Bitcoin Trend Strategy
