Variational Bayesian Approach for Vector Autoregression Model Learning

EasyChair Preprint no. 4595

Abstract

The vector autoregressive (VAR) model is one of the cores of analyzing the structure of multivariate time series over time. VAR is becoming more and more popular with complex data structure and huge data size. However, at the same time, the traditional MCMC application on the VAR model encounters the problem of excessive calculation time. To overcome the problem, we proposed the variational Bayesian Method for the VAR data analysis. The performance of the proposed method is illustrated via simulation studies.

Keyphrases: VAR model, variable selection, Variational Bayes

@Booklet{EasyChair:4595,
  author = {Wei-Ting Lai and Ray-Bing Chen},
  title = {Variational Bayesian Approach for Vector Autoregression Model Learning},
  howpublished = {EasyChair Preprint no. 4595},

  year = {EasyChair, 2020}}