The Bayesian quantile regression and rough set classification: Taiwanese satisfaction level analysis
Abstract
Purpose
In this paper, the authors aim to propose to find the variables that affect the Taiwanese people’s satisfaction level of the general public with the government.
Design/methodology/approach
The authors intend to utilize the Bayesian quantile regression to explore the variables that affect the satisfaction of the general public at specific quantiles of Taiwanese satisfaction with the government and rough set classification to explore key variables related to the satisfaction level. Then they make the comparison of the classification among the two methods to obtain the performance of the classification.
Findings
The experiment result shows the major factors which have the positive relationship with the people who have higher satisfaction level with the central government. These factors include satisfaction with the uncorrupted performance of the central government; the evaluation of household’s economic condition one year after the present time; the satisfaction with the Taiwanese central government’s measures on food safety and the satisfaction with the 12 years primary education reform.
Originality/value
The study’s originality hinges on the application of Bayesian quantile regression and rough set classification to the analysis of the Taiwanese satisfaction with the government. It offers more insights on the key variables related to different satisfaction level and the classification performance between the two methods.
Keywords
Acknowledgements
The authors gratefully acknowledge the editors and two anonymous referees for their insights and comments. The authors acknowledge the financial support for the Macau Foundation Research Fund.
Citation
Wu, S. and Guo, J. (2017), "The Bayesian quantile regression and rough set classification: Taiwanese satisfaction level analysis", Kybernetes, Vol. 46 No. 7, pp. 1262-1274. https://doi.org/10.1108/K-06-2016-0124
Publisher
:Emerald Publishing Limited
Copyright © 2017, Emerald Publishing Limited