Research Title |
The Modified Boxplot for Outlier Detection |
Date of Distribution |
18 November 2018 |
Conference |
Title of the Conference |
INTERNATIONAL CONFERENCE ON APPLIED STATISTICS (ICAS) 2018 |
Organiser |
คณะวิทยาศาสตร์และเทคโนโลยี มหาวิทยาลัยสงขลานครินทร์ วิทยาเขตปัตตานี ร่วมกับ สมาคมสถิติแห่งประเทศไทย |
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 |
121-125 |
Editors/edition/publisher |
|
Abstract |
The aim of this study is to propose a new modified boxplot for outlier detection based on symmetry and skewed data which is called the MK boxplot.
This MK boxplot is modified from Kimber’s boxplot by using the ratio of lower split interquartile range and upper split interquartile range into the
fences of the boxplot. The performance of the boxplot is evaluated by the mean percentage of detected outliers in three cases of simulated data
(truncated, uncontaminated and contaminated data) and real data. Furthermore, the existing boxplots for outlier detection are used to make a
comparison with the MK boxplot as well. The results from simulated data show that the MK boxplot performs well for symmetric and skewed data
when sample size is greater than 30. However, the MK boxplot has better performance than the others for skewed data. Moreover, when the MK
boxplot is applied to the real data, it efficiently detects outliers as the shape of real data. |
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
|
|