Deep functional factor models: forecasting high-dimensional functional time series via Bayesian nonparametric factorization

Liu, Yirui, Qiao, Xinghao ORCID: 0000-0002-6546-6595, Pei, Yulong and Wang, Liying (2024) Deep functional factor models: forecasting high-dimensional functional time series via Bayesian nonparametric factorization. Proceedings of Machine Learning Research, 235. pp. 31709-31727. ISSN 2640-3498

Text (liu24aw) - Published Version Download (1MB)

• Scopus publication

Abstract

This paper introduces the Deep Functional Factor Model (DF2M), a Bayesian nonparametric model designed for analysis of high-dimensional functional time series. DF2M is built upon the Indian Buffet Process and the multi-task Gaussian Process, incorporating a deep kernel function that captures non-Markovian and nonlinear temporal dynamics. Unlike many black-box deep learning models, DF2M offers an explainable approach to utilizing neural networks by constructing a factor model and integrating deep neural networks within the kernel function. Additionally, we develop a computationally efficient variational inference algorithm to infer DF2M. Empirical results from four real-world datasets demonstrate that DF2M provides better explainability and superior predictive accuracy compared to conventional deep learning models for high-dimensional functional time series.

Item Type:	Article
Additional Information:	© 2024 The Author(s)
Divisions:	LSE
Subjects:	Q Science > QA Mathematics H Social Sciences > HA Statistics
Date Deposited:	01 Oct 2024 15:27
Last Modified:	12 Dec 2024 04:30
URI:	http://eprints.lse.ac.uk/id/eprint/125587

Actions (login required)

View Item