|
Publication
|
Research Title |
Estimating the Population Coefficient of Quartile Variation for Bootstrap Confidence Intervals |
Date of Distribution |
24 October 2018 |
Conference |
Title of the Conference |
INTERNATIONAL CONFERENCE ON APPLIED STATISTICS (ICAS) 2018 |
Organiser |
Faculty of Science and Technology, Prince of Songkla University, Pattani Campus |
Conference Place |
Centra by Centara, Government Complex Hotel & Convention Centre Chaeng Watthana Bangkok |
Province/State |
กรุงเทพมหานคร |
Conference Date |
24 October 2018 |
To |
26 October 2018 |
Proceeding Paper |
Volume |
ุ- |
Issue |
- |
Page |
95-98 |
Editors/edition/publisher |
|
Abstract |
This research aims to study bootstrap confidence intervals for Coefficient of Quartile Variation (CQV) in 2 methods namely, Nonparametric Bootstrap (NB) and Parametric Bootstrap (PB) method when estimate the population coefficient of quartile variation from 5 quartiles method, there are Tukey, Moore and McCabe (M&M), Mendenhall and Sincich (M&S), Freund and Perles (F&P) and Minitab. The data case have Normal(4, 1), Lognormal(0, 1.5), Lognormal(0, 1), Lognormal(0, 0.75), Lognormal(0, 0.5) and Lognormal(0, 0.25) distribution and the sample sizes (n) are 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 19, 20, 25, 26, 29, 30, 35, 36, 39, 40, 45, 46, 49 and 50. The consideration performance of confidence interval is considered from coverage probability (CP) at confidence level of 0.95 and the average width of confidence interval (AW) by using Monte Carlo simulation with R program. The result found that confidence interval of NB method when estimate the population coefficient of quartile variation by F&P quartile method performs better than the other methods when the distributions are Normal(4, 1), Lognormal(0, 1), Lognormal(0, 0.75) and Lognormal(0, 0.25) distribution in various sizes of sample. For the confidence interval of PB method, it can confirm that the F&P method for estimate quartile is more effective than the other methods for Normal(4, 1), Lognormal(0, 0.5) and Lognormal(0, 0.25) distribution in many levels of n. And the performance of PB confidence interval based on Minitab quartile method has higher efficiency than the other methods for Lognormal(0, 1.5), Lognormal(0, 1) and Lognormal(0, 0.75) distribution in various sizes of sample. |
Author |
|
Peer Review Status |
มีผู้ประเมินอิสระ |
Level of Conference |
นานาชาติ |
Type of Proceeding |
Full paper |
Type of Presentation |
Oral |
Part of thesis |
true |
Presentation awarding |
false |
Attach file |
|
Citation |
0
|
|
|
|
|
|
|