Quantocracy: This is a curated mashup of quantitative trading links.
AutoAlpha: an Efficient Hierarchical Evolutionary Algorithm for Mining Alpha Factors in Quantitative Investment – Tianping Zhang, Yuanqi Li, Yifei Jin, Jian Li
The multi-factor model is a widely used model in quantitative investment. The success of a multi-factor model is largely determined by the effectiveness of the alpha factors used in the model. This paper proposes a new evolutionary algorithm called AutoAlpha to automatically generate effective formulaic alphas from massive stock datasets. Specifically, first we discover an inherent pattern of the formulaic alphas and propose a hierarchical structure to quickly locate the promising part of space for search. Then we propose a new Quality Diversity search based on the Principal Component Analysis (PCA-QD) to guide the search away from the well-explored space for more desirable results. Next, we utilize the warm start method and the replacement method to prevent the premature convergence problem. Based on the formulaic alphas we discover, we propose an ensemble learning-to-rank model for generating the portfolio. The backtests in the Chinese stock market and the comparisons with several baselines further demonstrate the effectiveness of AutoAlpha in mining formulaic alphas for quantitative trading.
Optimizing Return Forecasts: A Bayesian Intermediary Asset Pricing Approach – Ming Gao, Cong Zhang
In this study, we propose an innovative Bayesian method to estimate panel break model, using economically motivated priors derived from intermediary asset pricing models. Our approach enhances the panel break model by integrating financial frictions and merging cross-sectional and time-series data. This amalgamation facilitates the regime change identification, selection of return predictors, and the estimation of factor premia, bolstering the forecasting of equity returns. We benchmark our model against the leading Bayesian forecasting technique of Smith and Timmermann (2019), demonstrating superior performance via both simulation and empirical data. Our model underscores the importance of leveraging asset holdings data and integrating intermediary friction logic for accurately detecting real-time regime changes tied to significant market events. These advancements lead to marked improvements in out-of-sample performance, illustrated by substantial cumulative returns and a superior Sharpe ratio.
Machine Learning algorithms have been widely used and proven effective in financial markets. In this paper, we introduced a Machine Learning model set trained on the residual factors from the Fama-French three-factor model (Fama and French, 1992) to find significant alpha factors. To include more information that the linear factor models did not have, we used time series-based Machine Learning models, like tree-based models and Shallow Neural Networks with time series features. We used the predicted residual factor to construct quintile portfolios and found that it provided a significant alpha return even if style factors were controlled. It helped us find the latent information in the residual term and prove the lack of time series information in the factor models.
Mean-variance analysis is widely used in portfolio management to identify the best portfolio that makes an optimal trade-off between expected return and volatility. Yet, this method has its limitations, notably its vulnerability to estimation errors and its reliance on historical data. While shrinkage estimators and factor models have been introduced to improve estimation accuracy through bias-variance trade-offs, and the Black-Litterman model has been developed to integrate investor opinions, a unified framework combining three approaches has been lacking. Our study debuts a Bayesian blueprint that fuses shrinkage estimation with view inclusion, conceptualizing both as Bayesian updates. This model is then applied within the context of the Fama-French approach factor models, thereby integrating the advantages of each methodology. Finally, through a comprehensive empirical study in the US equity market spanning a decade, we show that the model outperforms both the simple 1/N portfolio and the optimal portfolios based on sample estimators.