The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Imputation method for missing data based on measure of property

특성도를 이용한 결측치 대체방법

  • Kim, Hyungju (Department of Biomedicine.Health Science, The Catholic University of Korea) ;
  • Kim, Dongjae (Department of Biomedicine.Health Science, The Catholic University of Korea)
  • 김형주 (가톨릭대학교 의생명.건강과학과) ;
  • 김동재 (가톨릭대학교 의생명.건강과학과)
  • Received : 2017.04.05
  • Accepted : 2017.05.05
  • Published : 2017.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

How to handle missing data is a main issue in clinical trials. We impute missing data based on missing data that follows a mechanism according to the intention-to-treat rule. However, using the right imputation method for missing data is very important because this supposition is unclear. We suggest a new imputation method for missing data using agreement and maintenance introduced by Kang and Kim (1997). We give an example and adapt a Monte Carlo simulation to compare the performance between the established method and the suggested method.

임상시험에서 어떻게 결측치를 다룰 것인가 하는 것은 큰 문제이다. 주로 주분석에서 사용하는 ITT원칙은 결측치가 어떠한 메커니즘을 따른다는 가정 하에 결측치를 대체 하지만 가정에 대한 타당성이 불확실한 문제가 있다. 즉, 올바른 결측치 대체방법은 매우 중요하다. 본 연구에서는 Kang과 Kim (1997)이 제안한 일치도와 유지도의 개념을 이용하여 새로운 결측치 대체방법을 제안하였다. 또한 실제자료를 이용하여 예제를 제시하고 Monte Carlo 모의실험을 통하여 기존방법과 대체 성능을 비교하였다.

Keywords

References

  1. Jo, B. and Kim, D. (2015). Comparison of single imputation methods in 22 cross-over design with missing observations, Korean Journal of Applied Statistics, 28, 529-540. https://doi.org/10.5351/KJAS.2015.28.3.529
  2. Kang, H. and Kim, B. (1997). A new measure of tracking in repeated measurement data, The Korean Communications in Statistics, 10, 189-201.
  3. Kang, S. (2013). Medical Statistics for New Medicine Development, Freeacademy, Seoul.
  4. Kim, D. (2009). Quantitative methods for missing value in the study of public administration - focused on the application of SPSS/MVA, Korean Comparative Government Review, 13, 177-196. https://doi.org/10.18397/kcgr.2009.13.1.177
  5. Lee, S. (2008). Conjugation plan of Proc MI, Industrial Science Research, 26, 35-41.
  6. Lee, S., An, J., and Kim, S. (2012). Comparison of imputation methods for GARCH model, Bulletin of the Natural Sciences, 26, 35-42.