Please use this identifier to cite or link to this item:
http://kmutnb-ir.kmutnb.ac.th/jspui/handle/123456789/555| Title: | Estimating confidence intervals for zero-inflated Rayleigh distributions การประมาณช่วงความเชื่อมั่นสำหรับการแจกแจงเรย์ลีที่มีค่าศูนย์ |
| Authors: | SASIPONG KIJSASON ศศิพงศ์ กิจสาสน SA-AAT NIWITPONG สอาด นิวิศพงศ์ King Mongkut's University of Technology North Bangkok มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ SUPARAT NIWITPONG สุภารัตน์ นิวิศพงศ์ snw@kmutnb.ac.th suparat.n@sci.kmutnb.ac.th,suparatn@kmutnb.ac.th |
| Keywords: | Resampling Generalized pivotal quantities Delta method Markov chain Monte Carlo Highest posterior density |
| Issue Date: | 6-Nov-2026 |
| Publisher: | King Mongkut's University of Technology North Bangkok |
| Abstract: | The zero-inflated two-parameter Rayleigh (ZITR) distribution is a mixture distribution that combines right-skewed continuous data with excess zero observations, where the occurrence of zeros follows a binomial distribution. This thesis aims to investigate parameter estimation for the ZITR distribution, with particular interest in the means, variances, and coefficients of variation (CV). Furthermore, the study extends the investigation to the difference and ratio of the mean, variance, and CV between two independent populations. For interval estimation of the parameters, several confidence interval methods were considered, including Bayesian, bootstrap, generalized confidence interval (GCI), and approximate normal (AN) methods. The performance of these methods was evaluated in terms of coverage probability (CP) and expected length (EL) through Monte Carlo simulation. A satisfactory method was expected to achieve a CP close to or greater than the nominal confidence level of 0.95 while providing the shortest expected length. In addition, real-world data were used to demonstrate the practical applicability of the proposed confidence interval methods. The applications included road traffic fatality data and road traffic fatality rates, illustrating the usefulness of the ZITR distribution for modeling data characterized by right skewness and excess zero observations. |
| URI: | http://kmutnb-ir.kmutnb.ac.th/jspui/handle/123456789/555 |
| Appears in Collections: | FACULTY OF APPLIED SCIENCE |
Files in This Item:
| File | Size | Format | |
|---|---|---|---|
| s6504051921519.pdf | 4.68 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.