Machine Learning Risk Models

Journal of Risk & Control 6(1) (2019) 37-64

26 Pages Posted: 8 Jan 2019 Last revised: 10 Apr 2019

See all articles by Zura Kakushadze

Zura Kakushadze

Quantigic Solutions LLC; Free University of Tbilisi

Willie Yu

Duke-NUS Medical School - Centre for Computational Biology

Date Written: January 1, 2019


We give an explicit algorithm and source code for constructing risk models based on machine learning techniques. The resultant covariance matrices are not factor models. Based on empirical backtests, we compare the performance of these machine learning risk models to other constructions, including statistical risk models, risk models based on fundamental industry classifications, and also those utilizing multilevel clustering based industry classifications.

Keywords: machine learning, risk model, clustering, k-means, statistical risk models, covariance, correlation, variance, cluster number, risk factor, optimization, regression, mean-reversion, factor loadings, principal component, industry classification, quant, trading, dollar-neutral, alpha, signal, backtest

JEL Classification: G00, G10, G11, G12, G23

Suggested Citation

Kakushadze, Zura and Yu, Willie, Machine Learning Risk Models (January 1, 2019). Journal of Risk & Control 6(1) (2019) 37-64, Available at SSRN: or

Zura Kakushadze (Contact Author)

Quantigic Solutions LLC ( email )

680 E Main St #543
Stamford, CT 06901
United States
6462210440 (Phone)
6467923264 (Fax)


Free University of Tbilisi ( email )

Business School and School of Physics
240, David Agmashenebeli Alley
Tbilisi, 0159

Willie Yu

Duke-NUS Medical School - Centre for Computational Biology ( email )

8 College Road
Singapore, 169857

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
PlumX Metrics