The Variational Bayesian Inference for Network Autoregression Models

40 Pages Posted: 23 Feb 2021

See all articles by Wei-Ting Lai

Wei-Ting Lai

Department of Statistics, National Cheng Kung University

Ray-Bing Chen

National Cheng Kung University

Ying Chen

National University of Singapore (NUS) - Department of Mathematics

Thorsten Koch

Technische Universitat Berlin - Software and Algorithms for Discrete Optimization

Date Written: February 18, 2021

Abstract

We develop a variational Bayesian (VB) approach for estimating large-scale dynamic network models in the network autoregression framework. The VB approach allows for the automatic identification of the dynamic structure of such a model and obtains a direct approximation of the posterior density. Compared to Markov Chain Monte Carlo (MCMC) based sampling approaches, the VB approach achieves enhanced computational efficiency without sacrificing estimation accuracy. In the simulation study conducted here, the proposed VB approach detects various types of proper active structures for dynamic network models. Compared to the alternative approach, the proposed method achieves similar or better accuracy, and its computational time is halved. In a real data analysis scenario of day-ahead natural gas flow prediction in the German gas transmission network with 51 nodes between October 2013 and September 2015, the VB approach delivers promising forecasting accuracy along with clearly detected structures in terms of dynamic dependence.

Keywords: Dynamic network; EM algorithm; MCMC algorithm; Vector autoregression

JEL Classification: C11;C32

Suggested Citation

Lai, Wei-Ting and Chen, Ray-Bing and Chen, Ying and Koch, Thorsten, The Variational Bayesian Inference for Network Autoregression Models (February 18, 2021). Available at SSRN: https://ssrn.com/abstract=3787923 or http://dx.doi.org/10.2139/ssrn.3787923

Wei-Ting Lai (Contact Author)

Department of Statistics, National Cheng Kung University ( email )

No.1, University Road
Tainan
Taiwan

Ray-Bing Chen

National Cheng Kung University

No.1, University Road
Tainan
Taiwan

Ying Chen

National University of Singapore (NUS) - Department of Mathematics ( email )

119076
Singapore

Thorsten Koch

Technische Universitat Berlin - Software and Algorithms for Discrete Optimization ( email )

Berlin
Germany

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
32
Abstract Views
205
PlumX Metrics