보조 정보에 의한 이중적 로버스트 대체법 = Doubly Robust Imputation Using Auxiliary Information

다국어 초록 (Multilingual Abstract)

Ratio and regression imputations depend on the model of a survey variable and the relation between the survey variable and auxiliary variables. If the model is not true, the unbiasedness of the estimator using the ratio or regression imputation cannot be guaranteed. In this paper, we develop the doubly robust imputation, which satisfies the approximate unbiasedness of the estimator, whether the model assumption is valid or not. The proposed imputation increases the efficiency of estimation by using the population information of the auxiliary variables. The simulation study establishes the theoretical results of this paper.

참고문헌 (Reference)

1 Rao, J. N. K, "Variance estimation under two-phase sampling with application toimputation for missing data" 82 : 453-460, 1995

2 Isaki, C. T, "Survey design under the regression superpopulation model" 77 : 89-96, 1982

3 Groves, R., "Survey Nonresponse" Wiley 2002

4 Rao,J.N.K, "On variance estimation with imputed survey data" 91 : 499-506, 1996

5 Rosenbaum,P.R, "Model-based direct adjustment" 82 : 387-394, 1987

6 Kim, J. K, "Imputation using response probability" 34 : 171-182, 2006

7 Cao, W., "Improving efficiency and robustness of the doubly robustestimator for a population mean with incomplete data" 96 : 723-734, 2009

8 Robins, J. M., "Estimation of regression coefficients when someregressors are not always observed" 89 : 846-866, 1994

9 Qin, J, "Efficient and doubly robust imputation for covariate-dependentmissing responses" 103 : 797-810, 2008

10 Kalton,G, "Compensating for Missing Survey Data, Institute for Social Research" University of Michigan 1983

11 Fay,R.E, "A design-based perspective on missing data variance, The ASA Proceedings of Bureauof the Census Annual Research Conference" 429-440, 1991

12 Carpenter, J. R, "A comparison of multiple imputation and doubly robustestimation for analyses with missing data" 169 : 571-584, 2006