Machine Learning and Factor-Based Portfolio Optimization

Thomas Conlon, John Cotter and Iason Kynigakis
Additional contact information
Iason Kynigakis: Smurfit Graduate Business School, University College Dublin

No 202111, Working Papers from Geary Institute, University College Dublin

Abstract: We examine machine learning and factor-based portfolio optimization. We find that factors based on autoencoder neural networks exhibit a weaker relationship with commonly used characteristic-sorted portfolios than popular dimensionality reduction techniques. Machine learning methods also lead to covariance and portfolio weight structures that diverge from simpler estimators. Minimum-variance portfolios using latent factors derived from autoencoders and sparse methods outperform simpler benchmarks in terms of risk minimization. These effects are amplified for investors with an increased sensitivity to risk-adjusted returns, during high volatility periods or when accounting for tail risk. Covariance matrices with a time-varying error component improve portfolio performance at a cost of higher turnover.

Keywords: Autoencoder; Covariance matrix; Dimensionality reduction; Factor models; Machine learning; Minimum-variance; Principal component analysis; Partial least squares; Portfolio optimization; Sparse principal component analysis; Sparse partial least squares (search for similar items in EconPapers)
JEL-codes: C38 C4 C45 C5 C58 G1 G11 (search for similar items in EconPapers)
Pages: 72 pages
Date: 2021-03-11
New Economics Papers: this item is included in nep-big, nep-cmp, nep-isf, nep-ore and nep-rmg
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.ucd.ie/geary/static/publications/workingpapers/gearywp202111.pdf First version, 2021 (application/pdf)

Related works:
Working Paper: Machine Learning and Factor-Based Portfolio Optimization (2021)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ucd:wpaper:202111

Access Statistics for this paper

More papers in Working Papers from Geary Institute, University College Dublin Contact information at EDIRC.
Bibliographic data for series maintained by Geary Tech ().