Abstract: We provide a comprehensive analysis of  timing success for equity risk factors. Our analysis covers over 300 risk factors (factor zoo) and a high-dimensional set of predictors. The performance of almost all factor groups can be improved through timing, especially profitability and value factors. Past factor returns and volatility stand out as the most successful individual predictors of factor returns. However, both are dominated by aggregating many predictors using partial least squares. The median improvement of a timed vs. untimed factor is about 2% p.a. A timed multifactor portfolio leads to an 8.6% increase in annualized return relative to its naively timed counterpart.

Presentations: Alpine Finance Summit (2024), Citigroup (2024), Arrowstreet Capital(2024), AFA (2024), Kepos Capital, DGF(2023), NFA (2023), CICF (2023), SGF (2023), AWG (2022)

Media coverage: institutionalinvestor.com (04/23), institutional-money.com (03/23) 

Abstract: This paper demonstrates that portfolio performance can be substantially enhanced by simultaneously utilizing historical factor return volatilities and option-derived market volatilities to optimize factor exposures. The improvements are particularly pronounced in regimes where option-implied market returns exhibit high volatility and right-skewness. Further gains in risk-adjusted portfolio returns are achieved by estimating model parameters separately for different regimes. Qualitatively similar results are obtained when all parameters are estimated strictly out-of-sample. These findings are not limited to a specific set of factors; comparable enhancements are observed when employing principal components derived from a broad set of factors.

Media coverage: 2nd Place Quantpedia Awards 2024 

Performance in different implied volatility and skewness regimes

Abstract: While the academic literature focuses on beta exposure, most practitioners apply characteristics-based scorings to obtain factor portfolios. This paper explores how firm-level characteristics can be combined for optimal factor portfolios. Portfolios encompassing multiple factors are less volatile and have higher after-cost returns than the market or single factor portfolios. We also demonstrate that buy- and sell-thresholds play a critical role in shaping portfolio return, risk, and turnover preferences. Our empirical findings reveal optimal weights for the combination of individual factors, though we acknowledge the 1/N portfolio as a challenging benchmark. 

Presentations:  APEF (2022), University of Liechtenstein (2022), Asian Finance Association (2021), ATINER (2021)

Media coverage: alphaarchitect.com (06/22)