Display options
Cite
Paik MC, Wang C. HANDLING MISSING DATA BY DELETING COMPLETELY OBSERVED RECORDS. J Stat Plan Inference. 2009;139(7):2341-2350doi: 10.1016/j.jspi.2008.10.024.
Paik, M. C., & Wang, C. (2009). HANDLING MISSING DATA BY DELETING COMPLETELY OBSERVED RECORDS. Journal of statistical planning and inference, 139(7), 2341-2350. https://doi.org/10.1016/j.jspi.2008.10.024
Paik, Myunghee Cho, and Wang, Cuiling. "HANDLING MISSING DATA BY DELETING COMPLETELY OBSERVED RECORDS." Journal of statistical planning and inference vol. 139,7 (2009): 2341-2350. doi: https://doi.org/10.1016/j.jspi.2008.10.024
Paik MC, Wang C. HANDLING MISSING DATA BY DELETING COMPLETELY OBSERVED RECORDS. J Stat Plan Inference. 2009 Jul 01;139(7):2341-2350. doi: 10.1016/j.jspi.2008.10.024. PMID: 20160863; PMCID: PMC2674251.
Copy Download .nbib Format: NLM
Share it on

J Stat Plan Inference. 2009 Jul 01;139(7):2341-2350. doi: 10.1016/j.jspi.2008.10.024.

HANDLING MISSING DATA BY DELETING COMPLETELY OBSERVED RECORDS.

Journal of statistical planning and inference

Myunghee Cho Paik, Cuiling Wang

Affiliations

  1. Department of Biostatistics, Mailman School of Public Health, Columbia University, 722 West 168 Street, New York City, N.Y. 10032, U.S.A.

PMID: 20160863 PMCID: PMC2674251 DOI: 10.1016/j.jspi.2008.10.024

Abstract

When data are missing, analyzing records that are completely observed may cause bias or inefficiency. Existing approaches in handling missing data include likelihood, imputation and inverse probability weighting. In this paper, we propose three estimators inspired by deleting some completely observed data in the regression setting. First, we generate artificial observation indicators that are independent of outcome given the observed data and draw inferences conditioning on the artificial observation indicators. Second, we propose a closely related weighting method. The proposed weighting method has more stable weights than those of the inverse probability weighting method (Zhao and Lipsitz, 1992). Third, we improve the efficiency of the proposed weighting estimator by subtracting the projection of the estimating function onto the nuisance tangent space. When data are missing completely at random, we show that the proposed estimators have asymptotic variances smaller than or equal to the variance of the estimator obtained from using completely observed records only. Asymptotic relative efficiency computation and simulation studies indicate that the proposed weighting estimators are more efficient than the inverse probability weighting estimators under wide range of practical situations especially when when the missingness proportion is large.

References

  1. Biometrics. 1996 Dec;52(4):1165-82 - PubMed
  2. Stat Med. 1997 Jan 15-Feb 15;16(1-3):5-19 - PubMed
  3. Biometrics. 1999 Mar;55(1):214-23 - PubMed
  4. Stat Med. 1988 Jan-Feb;7(1-2):305-15 - PubMed
  5. Stat Med. 1992 Apr;11(6):769-82 - PubMed
  6. Biometrics. 1999 Sep;55(3):984-9 - PubMed

Publication Types

Grant support