Araştırma Makalesi
BibTex RIS Kaynak Göster

Rastlantısal Olmayan Kayıp Veri Varlığında Seçim Modelleri ile bir Duyarlılık Analizi Uygulaması

Yıl 2017, Cilt: 10 Sayı: 2, 76 - 85, 30.12.2017

Öz

Bu çalışmadaki amaç, bağımlı değişkende rastlantısal olmayan kayıp veriler olduğunda, nasıl bir istatistiksel modelleme stratejisi izlenebileceğini açıklamak ve regresyon parametrelerinin farklı kayıp veri mekanizmaları varsayımlarına ne kadar duyarlı olabileceğini göstermektir. Bu amaç doğrultusunda, bir hanehalkı araştırmasından elde edilen veriler kullanılmış ve eğitim seviyesinin gelir düzeyi üzerindeki etkisi, doğrusal regresyon modeli kullanılarak, farklı kayıp veri mekanizmaları varsayımları altında ölçülmüştür. Analiz için Bayesci tahmin yöntemleri kullanılarak seçim modelleri yardımı ile, regresyon modeli ve kayıp veri modeli bileşik olarak modellenmiştir. Kayıp veri modelinin parametreleri değiştirilerek duyarlılık analizi yapılmış ve farklı kayıp veri mekanizmaları altında tahmin edilen regresyon katsayılarında ciddi farklılıklar görülmüştür.

Kaynakça

  • [1] P.D Allison, 2002, Missing Data. Sage Publications Inc, California.
  • 2] P. Diggle, M. G. Kenward, 1994, Informative drop-out in longitudinal data-analysis, Applied Statistics, 43, 49–93.
  • [3] A. Gelman, J.B. Carlin, H.S. Stern, D. B. Rubin, 2004, Bayesian Data Analysis, Chapman & Hall, Florida.
  • [4] A. Gelman, D. Rubin, 1992, Inference from Iterative Simulation using Multiple Sequences, Statistical Science, 7, 457-511.
  • [5] J. Geweke, 1992, Evaluating the Accuracy of Sampling-Based Approaches to Calculating Posterior Moments, J. M. Bernardo, J. M. Berger, A. P. Dawiv, A. F. Smith, Bayesian Statistics, (s. 196-193), University Press, Oxford.
  • [6] J.J. Heckman, 1979, Sample Selection Bias as a Specification Error, Econometrica, 47, 153-161
  • [7] F.T. Juster, J.P. Smith, 1997, Improving the Quality of Economic Data: Lessons from the HRS and AHEAD, Journal of the American Statistical Association, 92, 1268-1278.
  • [8] D. J. Lunn, A. Thomas, N. Best, D. Spiegelhalter, 2000, WinBUGS – A Bayesian modelling framework: Concepts, structure, and extensibility. Statistics and Computing, 10, 325–337.
  • [9] R. Little, 2008, Selection and Pattern-Mixture Models, G. Fitmaurice ve diğerleri, Advances in Longitudinal Data Analysis (18. Bölüm). CRC Press, London.
  • [10] B. Mandal, E. A. Stasny, 2004, Imputing Missing Income Data and Weighting Data with Imputed Income, Proceedings of the 2004 Joint Statistical Meetings, American Statistical Association, 3962- 3980.
  • [11] G. Molenberghs, M.G. Kenward, 2007, Missing Data in Clinical Studies. John Wiley and Sons, Chichester, UK.
  • [12] J. Moore, L. Stinson, E. Welniak, 1999, Income Reporting in Surveys Cognitive Issues and Measurement Error, M. D. Sirkin ve diğerleri (Ed.), Cognition and Survey Research (10. Bölüm). John Wiley & Sons, New York.
  • [13] C. Nicoletti, F. Peracchi, F. Foliano, 2011, Estimating income poverty in the presence of missing data and measurement error, Journal of Business & Economic Statistics, 29, 61–72. [14] T. E. Raghunathan, 2004, What do we do with missing data? Some options for analysis of incomplete data, Annual Review of Public Health, 25, 88–117.
  • [15] D. B. Rubin, 1976, Inference and Missing Data, Biometrika, 63, 581-592.
  • [16] D.B. Rubin, 2004, Multiple imputation for Nonresponse in Surveys. John Wiley & Sons, New York.
  • [17] D.B. Rubin, F. Little, 2002, Statistical Analysis with Missing Data, (Ed.), John Wiley and Sons, New Jersey.
  • [18] J. A. Sterne, I. R. White., J. B. Carlin, M. Spratt, P. Royston, M. Kenward, et al., 2009, Multiple imputation for missing data in epidemiological and clinical research: Potential and pitfalls. British Medical Journal, 338, b2393.
  • [19] H. Thijs, G. Molenberghs, B. Michiels, G. Verbeke, D. Curran, 2002, Strategies to fit pattern-mixture models. Biostatistics, 3, 245-265

An Application of Sensitivity Analysis in the Presence of Non-random Missing Data Using Selection Models

Yıl 2017, Cilt: 10 Sayı: 2, 76 - 85, 30.12.2017

Öz

The aim of this research is to explain a statistical modelling strategy in the presence of non-random missing data, and thereby to indicate how sensitive would the regression parameter estimates be under different assumptions of missing data mechanisms. For this purpose, a data set from a household survey was used, and the effect of education on individuals’ income levels has been assessed under different assumptions of missing data mechanisms by using a linear regression model. The modelling framework that was used comprised both the regression model and a missing data model using Bayesian estimation techniques jointly. Sensitivity analysis was carried out each time by changing the parameters of missing data model. It has been found out that under different assumptions of missing data mechanisms the parameter estimates of the regression model altered significantly.

Kaynakça

  • [1] P.D Allison, 2002, Missing Data. Sage Publications Inc, California.
  • 2] P. Diggle, M. G. Kenward, 1994, Informative drop-out in longitudinal data-analysis, Applied Statistics, 43, 49–93.
  • [3] A. Gelman, J.B. Carlin, H.S. Stern, D. B. Rubin, 2004, Bayesian Data Analysis, Chapman & Hall, Florida.
  • [4] A. Gelman, D. Rubin, 1992, Inference from Iterative Simulation using Multiple Sequences, Statistical Science, 7, 457-511.
  • [5] J. Geweke, 1992, Evaluating the Accuracy of Sampling-Based Approaches to Calculating Posterior Moments, J. M. Bernardo, J. M. Berger, A. P. Dawiv, A. F. Smith, Bayesian Statistics, (s. 196-193), University Press, Oxford.
  • [6] J.J. Heckman, 1979, Sample Selection Bias as a Specification Error, Econometrica, 47, 153-161
  • [7] F.T. Juster, J.P. Smith, 1997, Improving the Quality of Economic Data: Lessons from the HRS and AHEAD, Journal of the American Statistical Association, 92, 1268-1278.
  • [8] D. J. Lunn, A. Thomas, N. Best, D. Spiegelhalter, 2000, WinBUGS – A Bayesian modelling framework: Concepts, structure, and extensibility. Statistics and Computing, 10, 325–337.
  • [9] R. Little, 2008, Selection and Pattern-Mixture Models, G. Fitmaurice ve diğerleri, Advances in Longitudinal Data Analysis (18. Bölüm). CRC Press, London.
  • [10] B. Mandal, E. A. Stasny, 2004, Imputing Missing Income Data and Weighting Data with Imputed Income, Proceedings of the 2004 Joint Statistical Meetings, American Statistical Association, 3962- 3980.
  • [11] G. Molenberghs, M.G. Kenward, 2007, Missing Data in Clinical Studies. John Wiley and Sons, Chichester, UK.
  • [12] J. Moore, L. Stinson, E. Welniak, 1999, Income Reporting in Surveys Cognitive Issues and Measurement Error, M. D. Sirkin ve diğerleri (Ed.), Cognition and Survey Research (10. Bölüm). John Wiley & Sons, New York.
  • [13] C. Nicoletti, F. Peracchi, F. Foliano, 2011, Estimating income poverty in the presence of missing data and measurement error, Journal of Business & Economic Statistics, 29, 61–72. [14] T. E. Raghunathan, 2004, What do we do with missing data? Some options for analysis of incomplete data, Annual Review of Public Health, 25, 88–117.
  • [15] D. B. Rubin, 1976, Inference and Missing Data, Biometrika, 63, 581-592.
  • [16] D.B. Rubin, 2004, Multiple imputation for Nonresponse in Surveys. John Wiley & Sons, New York.
  • [17] D.B. Rubin, F. Little, 2002, Statistical Analysis with Missing Data, (Ed.), John Wiley and Sons, New Jersey.
  • [18] J. A. Sterne, I. R. White., J. B. Carlin, M. Spratt, P. Royston, M. Kenward, et al., 2009, Multiple imputation for missing data in epidemiological and clinical research: Potential and pitfalls. British Medical Journal, 338, b2393.
  • [19] H. Thijs, G. Molenberghs, B. Michiels, G. Verbeke, D. Curran, 2002, Strategies to fit pattern-mixture models. Biostatistics, 3, 245-265
Toplam 18 adet kaynakça vardır.

Ayrıntılar

Birincil Dil Türkçe
Bölüm Makaleler
Yazarlar

Oya Kalaycıoğlu 0000-0003-2183-7080

Yayımlanma Tarihi 30 Aralık 2017
Yayımlandığı Sayı Yıl 2017 Cilt: 10 Sayı: 2

Kaynak Göster

IEEE O. Kalaycıoğlu, “Rastlantısal Olmayan Kayıp Veri Varlığında Seçim Modelleri ile bir Duyarlılık Analizi Uygulaması”, JSSA, c. 10, sy. 2, ss. 76–85, 2017.