×

A simulation study comparing multiple imputation methods for incomplete longitudinal ordinal data. (English) Zbl 1328.62333

Summary: Multiple imputation (MI) is now a reference solution for handling missing data. The default method for MI is the Multivariate Normal Imputation (MNI) algorithm that is based on the multivariate normal distribution. In the presence of longitudinal ordinal missing data, where the Gaussian assumption is no longer valid, application of the MNI method is questionable. This simulation study compares the performance of the MNI and ordinal imputation regression model for incomplete longitudinal ordinal data for situations covering various numbers of categories of the ordinal outcome, time occasions, sample sizes, rates of missingness, well-balanced, and skewed data.

MSC:

62H12 Estimation in multivariate analysis
62J12 Generalized linear models (logistic models)
65C10 Random number generation in numerical analysis
62-04 Software, source code, etc. for problems pertaining to statistics
92B15 General biostatistics
PDFBibTeX XMLCite
Full Text: DOI Link

References:

[1] Ake C., Rounding after multiple imputation with non-binary categorical covariates (2005)
[2] Allison P., Imputation of categorical variables with PROC MI (2005)
[3] DOI: 10.1016/j.csda.2007.04.020 · Zbl 1452.62795 · doi:10.1016/j.csda.2007.04.020
[4] DOI: 10.1080/10543406.2011.550096 · doi:10.1080/10543406.2011.550096
[5] Fay R.E., When are Inferences from Multiple Imputation Valid? (1992)
[6] DOI: 10.1080/00031305.1979.10482685 · doi:10.1080/00031305.1979.10482685
[7] DOI: 10.1007/s11121-007-0070-9 · doi:10.1007/s11121-007-0070-9
[8] DOI: 10.1198/0003130032314 · Zbl 1182.62002 · doi:10.1198/0003130032314
[9] DOI: 10.1016/j.cmpb.2011.06.003 · doi:10.1016/j.cmpb.2011.06.003
[10] DOI: 10.1016/S0167-9473(97)00030-3 · Zbl 0908.62016 · doi:10.1016/S0167-9473(97)00030-3
[11] DOI: 10.1093/biomet/73.1.13 · Zbl 0595.62110 · doi:10.1093/biomet/73.1.13
[12] DOI: 10.1002/sim.4780131106 · doi:10.1002/sim.4780131106
[13] DOI: 10.1080/01621459.1993.10594302 · doi:10.1080/01621459.1993.10594302
[14] DOI: 10.1080/01621459.1995.10476615 · doi:10.1080/01621459.1995.10476615
[15] Little R. J.A., Statistical Analysis with Missing Data (1987) · Zbl 0665.62004
[16] McCullagh P., Journal of the Royal Statistical Society, Series B 42 pp 109– (1980)
[17] Meng X.L., Statistical Science 9 pp 538– (1994)
[18] DOI: 10.1080/01621459.1995.10476494 · doi:10.1080/01621459.1995.10476494
[19] DOI: 10.1080/01621459.1995.10476493 · doi:10.1080/01621459.1995.10476493
[20] DOI: 10.1093/biomet/63.3.581 · Zbl 0344.62034 · doi:10.1093/biomet/63.3.581
[21] Rubin D.B., Proceedings of the Survey Research Methods Section, American Statistical Association pp 20– (1978)
[22] DOI: 10.1002/9780470316696 · doi:10.1002/9780470316696
[23] DOI: 10.1080/01621459.1996.10476908 · doi:10.1080/01621459.1996.10476908
[24] DOI: 10.1201/9781439821862 · doi:10.1201/9781439821862
[25] DOI: 10.1080/01621459.1987.10478458 · doi:10.1080/01621459.1987.10478458
[26] DOI: 10.1016/S0169-2607(98)00063-7 · doi:10.1016/S0169-2607(98)00063-7
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.