Statistics and Its Interface

Volume 15 (2022)

Number 1

Subset selection of double-threshold moving average models through the application of the Bayesian method

Pages: 51 – 61



Jinshan Liu (School of Financial Mathematics and Statistics, Guangdong University of Finance, Guangzhou, China)

Jiazhu Pan (Department of Mathematics and Statistics, University of Strathclyde, Glasgow, Scotland, United Kingdom)

Qiang Xia (Department of Applied Mathematics, South China Agricultural University, Guangzhou, China)

Ying Xiao (Department of Applied Mathematics, South China Agricultural University, Guangzhou, China)


The Bayesian method is firstly applied for the selection of the best subset for the double-threshold moving average (DTMA) model. The Markov chain Monte Carlo (MCMC) techniques and the stochastic search variable selection (SSVS) method are used to identify the best subset model from a very large number of possible models. Simulation experiments show that the proposed method is feasible and efficient, despite the complexity being increased by the large number of subsets, and the uncertainty of the threshold and delay variables. Our method is illustrated by real data analysis on the Yen-Dollar exchange rate.


Bayesian estimation, DTMA model, MCMC algorithm, SSVS, exchange rate

The authors’ names are listed here in alphabetical order.

This work was partially supported by the Major Research Plan of the National Natural Science Foundation of China (91746102), Ministry of Education in China Project of Humanities and Social Sciences (No. 17YJA910002), and the Open Research Fund of Key Laboratory of Advanced Theory and Application in Statistics and Data Science of East China Normal University, Ministry of Education.

Received 29 December 2019

Accepted 29 March 2021

Published 11 August 2021