search for




 

A study of discrete Weibull regression model with missing data
Journal of the Korean Data & Information Science Society 2019;30:11-22
Published online January 31, 2019;  https://doi.org/10.7465/jkdi.2019.30.1.11
© 2019 Korean Data and Information Science Society.

Hanna Yoo1

1Department of Computer Software, Busan University of Foreign Studies
Correspondence to: Assistant professor, Department of Computer Software, Busan University of Foreign Studies, 65 Geumsaem-ro 485 beon-gil, Geumjeong-gu, Busan, Korea. Email: pinkcan78@bufs.ac.kr
Received December 18, 2018; Revised January 7, 2019; Accepted January 11, 2019.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
Discrete Weibull regression model can be adapted to discrete count data with different type of dispersions. It can be adopted to various types of dispersion however it is not used widely in discrete data and there isn’t much research papers that deal with discrete Weibull regression model. In this paper, discrete Weibull regression model is adapted to data that has missing values. Single imputation method is used to impute the missing values. We analyzed the seventh Korea National Health and Nutrition Examination Survey (KNHANES Ⅶ), 2016 to assess the factors for 1 year hospital stay. We compare the results using discrete Weibull regression model with zero-inflated Poisson model and shown that discreteWeibull regression model provided better fit. We also performed simulation studies to show the accuracy of the discrete Weibull regression with using single imputation under various missing rates and sample size. Through simulation studies, it was shown that using imputation methods yield better results then deleting the missing values. Using imputation with discrete Weibull regression model to discrete data will increase the power and enables the wide applicability to various types of dispersion data.
Keywords : Discrete count data, discrete Weibull regression model, imputation method, KNHANES.