Machine Learning for Recession Prediction and Dynamic Asset Allocation

The Journal of Financial Data Science, Forthcoming

Posted: 28 Jan 2019 Last revised: 26 Jun 2019

See all articles by Alex James

Alex James

Paraconic Technologies US Inc.

Yaser Abu-Mostafa

California Institute of Technology

Xiao Qiao

Paraconic Technologies US Inc.

Date Written: December 21, 2018

Abstract

We introduce a novel application of support vector machines (SVM), an important machine learning algorithm, to determine the beginning and end of recessions in real time. Nowcasting, forecasting a condition in the present time because the full information will not be available until later, is key for recessions, which are only determined months after the fact. We show that SVM has excellent predictive performance for this task, capturing all six recessions from 1973 to 2018 and providing the signal with minimal delay. We take advantage of the timeliness of SVM signals to test dynamic asset allocation between stocks and bonds. A dynamic risk budgeting approach using SVM outputs appears superior to an equal-risk contribution portfolio, improving the average returns by 85 bps per annum without increased tail risk.

Keywords: Asset Allocation, Forecasting, Macroeconomics, Recessions, Machine Learning, Business Cycle, SVM

JEL Classification: C14, C53, E32, E37

Suggested Citation

James, Alex and Abu-Mostafa, Yaser and Qiao, Xiao, Machine Learning for Recession Prediction and Dynamic Asset Allocation (December 21, 2018). The Journal of Financial Data Science, Forthcoming. Available at SSRN: https://ssrn.com/abstract=3316917 or http://dx.doi.org/10.2139/ssrn.3316917

Alex James

Paraconic Technologies US Inc. ( email )

New York, NY
United States

Yaser Abu-Mostafa

California Institute of Technology ( email )

Pasadena, CA 91125
United States

Xiao Qiao (Contact Author)

Paraconic Technologies US Inc. ( email )

New York, NY
United States

HOME PAGE: http://sites.google.com/site/xiaoqiao10/

Register to save articles to
your library

Register

Paper statistics

Abstract Views
1,342
PlumX Metrics