A Multiple Imputation Framework for Massive Multivariate Data of Different Variable Types: A Monte-Carlo Technique
- 2.8k Downloads
The purpose of this chapter is to build theoretical, algorithmic, and implementation-based components of a unified, general-purpose multiple imputation framework for intensive multivariate data sets that are collected via increasingly popular real-time data capture methods . Such data typically include all major types of variables that are incomplete due to planned missingness designs, which have been developed to reduce respondent burden and lower the cost associated with data collection. The imputation approach presented herein complements the methods available for incomplete data analysis via richer and more flexible modeling procedures, and can easily generalize to a variety of research areas that involve internet studies and processes that are designed to collect continuous streams of real-time data. Planned missingness designs are highly useful and will likely increase in popularity in the future. For this reason, the proposed multiple imputation framework represents an important and refined addition to the existing methods, and has potential to advance scientific knowledge and research in a meaningful way. Capability of accommodating many incomplete variables of different distributional nature, types, and dependence structures could be a contributing factor for better comprehending the operational characteristics of today’s massive data trends. It offers promising potential for building enhanced statistical computing infrastructure for education and research in the sense of providing principled, useful, general, and flexible set of computational tools for handling incomplete data.
KeywordsRandom Number Generation Count Variable Standard Normal Variable Threshold Concept Tetrachoric Correlation
- Amatya, A., & Demirtas, H. (2015b). Concurrent generation of ordinal and normal data with given correlation matrix and marginal distributions, R package OrdNor. http://CRAN.R-project.org/package=OrdNor.
- Amatya, A., & Demirtas, H. (2016a). Simultaneous generation of multivariate binary and normal variates, R package BinNor. http://CRAN.R-project.org/package=BinNor.
- Amatya, A., & Demirtas, H. (2016b). Generation of multivariate ordinal variates, R package MultiOrd. http://CRAN.R-project.org/package=MultiOrd.
- Amatya, A., & Demirtas, H. (2016c). Simultaneous generation of multivariate data with Poisson and normal marginals, R package PoisNor. http://CRAN.R-project.org/package=PoisNor.
- Demirtas, H. (2004). Simulation-driven inferences for multiply imputed longitudinal datasets. Statistica Neerlandica, 58, 466–482.Google Scholar
- Demirtas, H. (2014). Joint generation of binary and nonnormal continuous data. Journal of Biometrics and Biostatistics, 5, 1–9.Google Scholar
- Demirtas, H. (2017a). Concurrent generation of binary and nonnormal continuous data through fifth order power polynomials, Communications in Statistics—Simulation and Computation, 46, 344–357.Google Scholar
- Demirtas, H. (2017b). On accurate and precise generation of generalized Poisson variates, Communications in Statistics—Simulation and Computation, 46, 489–499.Google Scholar
- Demirtas, H., Hu, Y., & Allozi, R. (2016b). Data generation with Poisson, binary, ordinal and normal components, R package PoisBinOrdNor. https://cran.r-project.org/web/packages/PoisBinOrdNor.
- Demirtas, H., Nordgren, R., & Allozi, R. (2016c). Generation of up to four different types of variables, R package PoisBinOrdNonNor. https://cran.r-project.org/web/packages/PoisBinOrdNonNor.
- Demirtas, H., Shi, Y., & Allozi, R. (2016d). Simultaneous generation of count and continuous data, R package PoisNonNor. https://cran.r-project.org/web/packages/PoisNonNor.
- Demirtas, H., Wang, Y., & Allozi, R. (2016e). Concurrent generation of binary, ordinal and continuous data, R package BinOrdNonNor. https://cran.r-project.org/web/packages/BinOrdNonNor.
- Emrich, J. L., & Piedmonte, M. R. (1991). A method for generating high-dimensional multivariate binary variates. The American Statistician, 45, 302–304.Google Scholar
- Inan, G., & Demirtas, H. (2016a). Data generation with binary and continuous non-normal components, R package BinNonNor. https://cran.r-project.org/web/packages/BinNonNor.
- Inan, G., & Demirtas, H. (2016b). Data generation with Poisson, binary and ordinal components, R package PoisBinOrd. https://cran.r-project.org/web/packages/PoisBinOrd.
- Inan, G., & Demirtas, H. (2016c). Data generation with Poisson, binary and continuous components, R package PoisBinNonNor. https://cran.r-project.org/web/packages/PoisBinNonNor.
- Raghunathan, T. E., Lepkowski, J. M., van Hoewyk, J., & Solenberger, P. A. (2001). Multivariate technique for multiply imputing missing values using a sequence of regression models. Survey Methodology, 27, 85–95.Google Scholar
- Schafer, J. L. (1997). Analysis of incomplete multivariate data. London, UK: Chapman and Hall.Google Scholar