search for




 

Comparison of the performances in approximating sample influence function using empirical influence function
Journal of the Korean Data & Information Science Society 2022;33:209-22
Published online March 31, 2022;  https://doi.org/10.7465/jkdi.2022.33.2.209
© 2022 Korean Data and Information Science Society.

Hyun Seok Kang1 · A La Jeh2 · Honggie Kim3

1Ministry of Education
2Statistics Korea
3Department of Information and Statistics, Chungnam National University
Correspondence to: 1 Educational researcher, Ministry of Education, 408, Galmae-ro, Sejong, Korea.
2 Officer, Statistics Korea, 189, Cheongsa-ro, Seo-gu, Daejeon, Korea.
3 Professor, Department of Information and Statistics, Chungnam National University, 99, Daehak-ro, Yuseong-gu, Daejeon 34134, Korea. E-mail : honggiekim@cnu.ac.kr
Received November 29, 2021; Revised December 30, 2021; Accepted January 3, 2022.
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
This study is a follow-up study of Kang and Kim (2020). In this study, we consider a method to approximate the sample influence function for sample standard deviation by the empirical influence function. The validity of approximation equation is verified through a simulation of a random sample of size 300. Especially, through the comparison of performances among those two approximation methods which approximate sample influence function for sample standard deviation, we examine the advantages and limitations of each approximation method. This study seems to have its significance in proposing both a method which reduces errors in approximation of the empirical influence function and an effective and practical method evolved from previous research in which the sample influence function is directly approximated through the empirical influence function by a simple constant revision.
Keywords : Approximation, empirical influence function, influence function, outlier, sample influence function, sample standard deviation.