Imputaciones múltiples, herramienta para la estimación de datos faltantes en la modelación de regresión

• Autores: Luis Miguel Mejía Giraldo, Luis Fernando Restrepo Betancur
• Localización: Temas agrarios, ISSN-e 0122-7610, Vol. 24, Nº. 1, 2019, págs. 66-73
• Idioma: español
• Títulos paralelos:
  • Multiple Imputations, tool for the estimation of missing data in regression modeling
• Enlaces
  • Texto completo
• Resumen
  • español

    En los últimos años se ha apreciado un incremento en la investigación sobre problemas de datos faltantes, siendo la imputación múltiple una fundamental alternativa; donde los conjuntos de datos a menudo presentan complejidades que son actualmente difíciles de manejar de manera apropiada en el marco de probabilidad, pero relativamente simples de tratar con imputación; por esto, el presente artículo describe una serie de aspectos prácticos para aplicar dicha metodología en el caso de la modelación de captura de carbono para Colombia, con base en las bases de datos del Banco Mundial incluyendo datos faltantes alcanzando R2 de 79,30%, resaltándose que al momento de estimar dichos datos y recalcular el modelo respectivo se evidencia un mayor R2, siendo del 94,79%, lo cual evidencia una mejora sustancial del respectivo modelo de regresión lineal múltiple como tal.

  • English

    In recent years there has been an increase in research on missing data problems, with multiple imputation being a fundamental alternative; where data sets often present complexities that are currently difficult to manage appropriately in the probability framework, but relatively simple to deal with imputation; For this reason, this article describes a series of practical aspects to apply this methodology in the case of carbon capture modeling for Colombia, based on the World Bank databases including missing data reaching R2 of 79.2988%, highlighting that when estimating said data and recalculating the respective model, a greater R2 is evidenced, being of 94.76901%, which evidences a substantial improvement of the respective multiple linear regression model as such.

• Referencias bibliográficas
  • Allison, P. 2002. Missing data. Newbury Park, CA: Sage. Asparouhov, T., & Muthen, B. (2010). Multiple imputation with Mplus. Retrieved...
  • Carpenter, J. and Kenward, M. 2013. Multiple imputation and its application. West Sussex, UK: Wiley.
  • Collins, L., Schafer, J. and Kam, C. 2001. A comparison of inclusive and restrictive strategies in modern missing data procedures. Psychological...
  • Enders, C. 2010. Applied missing data analysis. New York: Guilford Press.
  • Gelman, A., Carlin, J., Stern, H., Dunson, D., Vehtari, A. and Rubin, D. 2014. Bayesian data analysis (3rd ed.). Boca Raton, FL: CRC Press.
  • Graham, J. 2012. Missing data: Analysis and design. New York: Springer.
  • Jelicic, H., Phelps, E. and Lerner, R. 2009. Use of missing data methods in longitudinal studies: The persistence of bad practices in developmental...
  • Jimenez, J. and Mejía L. 2014. Multivariate Stochastic Analysis CO2 emission factor for carbon sequestration and sustainable development for...
  • Little, R. and Rubin, D. 2002.Statistical analysis with missing data. Hoboken, NJ: Wiley.
  • Meng, X.1994. Multiple-imputation inferences with uncongenial sources of input. Statistical Science, 9, 538e558.
  • Peugh, J. and Enders, C. 2004. Missing data in educational research: A review of reporting practices and suggestions for improvement. Review...
  • Raghunathan, T., Lepkowski, J., Van Hoewyk, J. and Solenberger, P. 2001. A multivariate technique for multiply imputing missing values using...
  • Raykov, T. 2011. On Testability of missing data mechanisms in incomplete datasets. Structural Equation Modeling: A Multidisciplinary Journal,...
  • Rubin, D. 1976. Inference and missing data. Biometrika, 63, 581e592.
  • Rubin, D. 1987. Multiple imputation for nonresponse in surveys. Hoboken, New Jersey: Wiley.
  • Rubin, D. 1996. Multiple imputation after 18þ years. Journal of the American Statistical Association, 91, 473e489.
  • Schafer, J. 1997. Analysis of incomplete multivariate data. New York: Chapman &Hall.
  • Schafer, J. 2003. Multiple imputation in multivariate problems when the imputation and analysis models differ. Statistica Neerlandica, 57,...
  • Schafer, J. and Graham, J. 2002. Missing data: Our view of the state of the art. Psychological Methods, 7, 147e177.
  • Van Buuren, S. 2012. Flexible imputation of missing data. New York: Chapman &Hall.
  • Van Buuren, S., Brand, J. Groothuis-Oudshoorn, C. and Rubin, D. 2006. Fully conditional specification in multivariate imputation. Journal...
  • Widaman, K. 2006. Missing data: What to do with or without them. Monographs of the Society for Research in Child Development, 71, 42e64.
  • Wood, A., White, I. and Thompson, S. 2004. Are missing outcome data adequately handled? A review of published randomized controlled trials...