DOI QR코드

DOI QR Code

Multiple Imputation Reducing Outlier Effect using Weight Adjustment Methods

가중치 보정을 이용한 다중대체법

  • Kim, Jin-Young (Department of Statistics, Hankuk University of Foreign Studies) ;
  • Shin, Key-Il (Department of Statistics, Hankuk University of Foreign Studies)
  • 김진영 (한국외국어대학교 통계학과) ;
  • 신기일 (한국외국어대학교 통계학과)
  • Received : 2013.04.27
  • Accepted : 2013.06.18
  • Published : 2013.08.31

Abstract

Imputation is a commonly used method to handle missing survey data. The performance of the imputation method is influenced by various factors, especially an outlier. The removal of the outlier in a data set is a simple and effective approach to reduce the effect of an outlier. In this paper in order to improve the precision of multiple imputation, we study a imputation method which reduces the effect of outlier using various weight adjustment methods that include the removal of an outlier method. The regression method in PROC/MI in SAS is used for multiple imputation and the obtained final adjusted weight is used as a weight variable to obtain the imputed values. Simulation studies compared the performance of various weight adjustment methods and Monthly Labor Statistic data is used for real data analysis.

다중 대체법은 표본조사에서 결측값이 발생하였을 때 가장 흔히 사용하는 방법이다. 이 방법은 여러 요인에 의해 그 성능이 좌우되며 특히 이상점의 영향을 많이 받는다. 본 연구에서는 가중치 보정법을 이용하여 이상점의 영향력을 줄여 다중 대체법의 성능을 향상시키는 방법을 연구하였다. 가중치 보정법을 이용하여 얻어진 최종 가중치를 다중대체에 사용하였으며 SAS의 PROC MI가 다중 대체를 위해 사용되었다. 모의실험과 매월노동통계 자료를 이용한 실제 자료 분석을 통하여 제안된 방법의 우수성을 확인하였다.

Keywords

References

  1. Belcher, R. (2003). Application of the Hidiroglou-Berthelot method of outlier detection for periodic business surveys, SSC Annual Meeting, Proceeding of the Survey Method Section.
  2. Hidiroglou, M. A. and Berthelot, J.-M. (1986). Statistical editing and imputation for Periodic Business Surveys, Survey Methodology, 12, 73-83.
  3. Hidiroglou, M. A. Srinath, K. P. (1981). Some estimators of a population total from simple random samples constraining large units, Journal of Applied Statistics, 76, 690-695.
  4. Kim, J. (2006). Weight reduction method for outlier in survey sampling, The Korean Communications in Statistics, 13, 19-27. https://doi.org/10.5351/CKSS.2006.13.1.019
  5. Lee, H., Rancourt, E. and Sarndal, C.-E.(1995). Experiment with variance estimation from survey data with imputed value, Journal of Official Statistics, 10, 231-243.
  6. Lee, S.-J. and Shin, K.-I. (2008). A Study on the sensitivity of the BLS Methods, Communications of the Korean Statistical Society, 15, 843-858. https://doi.org/10.5351/CKSS.2008.15.6.843
  7. McCullough, M. and Pennington, T. L. (2009). Identifying outliers when creating an imputation based for the Quarterly Financial Report, JSM, Section on Survey Research Methods.
  8. Rubin, D. B. (1987). Multiple Imputation for Nonresponse in Survey, New York.

Cited by

  1. A Multiple Imputation for Reducing Outlier Effect vol.27, pp.7, 2014, https://doi.org/10.5351/KJAS.2014.27.7.1229