Likelihood-based quantile autoregressive distributed lag models and its applications

ABSTRACT

Time lag effect exists widely in the course of economic operation. Some economic variables are affected not only by various factors in the current period but also by various factors in the past and even their own past values. As a class of dynamical models, autoregressive distributed lag (ARDL) models are frequently used to conduct dynamic regression analysis. In this paper, we are interested in the quantile regression (QR) modeling of the ARDL model in a dynamic framework. By combining the working likelihood of asymmetric Laplace distribution (ALD) with the expectation–maximization (EM) algorithm into the considered ARDL model, the iterative weighted least square estimators (IWLSE) are derived. Some Monte Carlo simulations are implemented to evaluate the performance of the proposed estimation method. A dataset of the consumption of electricity by residential customers is analyzed to illustrate the application.

Keywords:

• Dynamic regression model
• QR analysis
• EM algorithm
• information criterion
• electricity consumption

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Acknowledgments

The authors thank reviewers for their constructive comments and valuable suggestions which have greatly improved the paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The work is partly supported by China Postdoctoral Science Foundation (No. 2017M610156), National Natural Science Foundation of China (Nos. 11501167, 71473280), Young academic leaders project of Henan University of Science and Technology (No. 13490008) and Major project of the National Social Science Foundation of China (No. 15ZDA009).