Taming data-driven probability distributions

30 Pages Posted: 27 Apr 2022 Last revised: 30 May 2024

See all articles by Jozef Baruník

Jozef Baruník

Charles University in Prague - Department of Economics; Institute of Information Theory and Automation, Prague

Lubos Hanus

Institute of Economic Studies, Charles University in Prague

Date Written: April 14, 2022

Abstract

We propose a deep learning approach to probabilistic forecasting of macroeconomic and financial time series. By allowing complex time series patterns to be learned from a data-rich environment, our approach is useful for decision making that depends on the uncertainty of a large number of economic outcomes. In particular, it is informative for agents facing asymmetric dependence of their loss on the outcomes of possibly non-Gaussian and non-linear variables. We demonstrate the usefulness of the proposed approach on two different datasets where a machine learns patterns from the data. First, we illustrate the gains in predicting stock return distributions that are heavy tailed and asymmetric. Second, we construct macroeconomic fan charts that reflect information from a high-dimensional dataset.

Keywords: Distributional forecasting, machine learning, deep learning, probability, economic time series

JEL Classification: C45, C53, E17, E37

Suggested Citation

Barunik, Jozef and Hanus, Luboš, Taming data-driven probability distributions (April 14, 2022). Available at SSRN: https://ssrn.com/abstract=4083719 or http://dx.doi.org/10.2139/ssrn.4083719

Jozef Barunik (Contact Author)

Charles University in Prague - Department of Economics ( email )

Opletalova 26
Prague 1, 110 00
Czech Republic

HOME PAGE: http://ies.fsv.cuni.cz/en/staff/barunik

Institute of Information Theory and Automation, Prague ( email )

Pod vodarenskou vezi 4
CZ-18208 Praha 8
Czech Republic

HOME PAGE: http://staff.utia.cas.cz/barunik/home.htm

Luboš Hanus

Institute of Economic Studies, Charles University in Prague ( email )

Opletalova 26
Praha 1, 11000
Czech Republic

HOME PAGE: http://ies.fsv.cuni.cz/en/staff/hanusl

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

Paper statistics

Downloads
149
Abstract Views
532
Rank
370,113
PlumX Metrics