Time‐series and Cross‐sectional Stock Return Forecasting: New Machine Learning Methods

Rapach, David E.; Zhou, Guofu

doi:10.1002/9781119751182.ch1

Cited by 19 publications

(4 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dong et al (2022) use 100 long-short "anomaly" portfolios to forecast the market return using a variety of forecasting strategies to implement shrinkage (more generally, see the recent survey by Rapach and Zhou (2022)). An emerging literature uses machine learning methods to forecast large panels of individual stock returns or portfolios, including Rapach and Zhou (2020), Kozak, Nagel, andSantosh (2020), Freyberger, Neuhierl, andWeber (2020), Gu, Kelly, andXiu (2020), andChen, Pelger, andZhu (2023) (also see the survey by Kelly and Xiu (2022)). Our paper offers theoretical justification for the successes of machine learning prediction documented in the asset pricing literature.…”

mentioning

confidence: 99%

The Virtue of Complexity in Return Prediction

KELLY,

MALAMUD,

ZHOU

2023

The Journal of Finance

View full text Add to dashboard Cite

Much of the extant literature predicts market returns with “simple” models that use only a few parameters. Contrary to conventional wisdom, we theoretically prove that simple models severely understate return predictability compared to “complex” models in which the number of parameters exceeds the number of observations. We empirically document the virtue of complexity in U.S. equity market return prediction. Our findings establish the rationale for modeling expected returns through machine learning.

show abstract

mentioning

confidence: 99%

The Virtue of Complexity in Return Prediction

KELLY,

MALAMUD,

ZHOU

2023

The Journal of Finance

View full text Add to dashboard Cite

show abstract

“…For the special cases that ρ = 0 and ρ = 1, the ENet reduces to the Ridge and LASSO approach, respectively. For simplicity, we choose the value of ρ as 0.5 to balance the L 1 and L 2 penalties, following Rapach and Zhou (2020) and Zhang, Wahab, et al (2023). The LASSO and ENet approach can be regarded as efficient combinations of predictors based on regularization and variable selection.…”

Section: Methodsmentioning

confidence: 99%

The predictability of carbon futures volatility: New evidence from the spillovers of fossil energy futures returns

Zhang,

Wang

et al. 2024

Journal of Futures Markets

View full text Add to dashboard Cite

In this paper, we find new evidence for the carbon futures volatility prediction by using the spillovers of fossil energy futures returns as a powerful predictor. The in‐sample results show that the spillovers have a significantly positive effect on carbon futures volatility. From the out‐of‐sample analysis with various loss functions, we find that fossil energy return spillovers significantly outperform the benchmark and show better forecasting performance than the competing models using dimension reduction, variable selection, and combination approaches. The predictive ability of the spillovers also holds in long‐term forecasting and does not derive from other carbon‐related variables. It can bring substantial economic gains in the portfolio exercise within carbon futures. Finally, we provide economic explanations on the predictive ability of the fossil energy return spillover by the channels of the carbon emission uncertainty and the investor sentiment on the warming climate.

show abstract

“…A notable example is Campbell and Thompson (2008). More recently, Rapach, Strauss, and Zhou (2010), Diebold and Shin (2019), and Rapach and Zhou (2020) have used ML techniques to produce forecast combinations for time series such as market returns or macroeconomic variables. Rapach et al (2019) use similar techniques to forecast industry-level portfolio returns.…”

Section: A Related Literaturementioning

confidence: 99%

Uncovering Sparsity and Heterogeneity in Firm-Level Return Predictability Using Machine Learning

Evgeniou¹,

Guecioueur

Prieto³

2022

J. Financ. Quant. Anal.

View full text Add to dashboard Cite

We develop an approach that combines the estimation of monthly firm-level expected returns with an assignment of firms to (possibly) latent groups, both based on observable characteristics, using machine learning principles with linear models. The best-performing methods are flexible two-stage sparse models that capture group-membership predictive relationships. Portfolios formed to exploit such group-varying predictions based on a parsimonious set of characteristics deliver economically meaningful returns with low turnover. We propose statistical tests based on nonparametric bootstrapping for our results, and detail how different characteristics may matter for different groups of firms, making comparisons to the existing literature.

show abstract

Time‐series and Cross‐sectional Stock Return Forecasting: New Machine Learning Methods

Cited by 19 publications

References 31 publications

The Virtue of Complexity in Return Prediction

The Virtue of Complexity in Return Prediction

The predictability of carbon futures volatility: New evidence from the spillovers of fossil energy futures returns

Uncovering Sparsity and Heterogeneity in Firm-Level Return Predictability Using Machine Learning

Contact Info

Product

Resources

About