Journal of the Korean Data and Information Science Society

The Korean Data and Information Science Society (한국데이터정보과학회)
권호 표지

Comparative Study on Imputation Procedures in Exponential Regression Model with missing values

  • Park, Young-Sool (Dept. of Information Statistics, Kwandong University) ;
  • Kim, Soon-Kwi (Dept. of Information Statistics, Kangnung National University)
  • Published : 2003.05.31
Download PDF
⟨ Previous Next ⟩

Abstract

A data set having missing observations is often completed by using imputed values. In this paper, performances and accuracy of five imputation procedures are evaluated when missing values exist only on the response variable in the exponential regression model. Our simulation results show that adjusted exponential regression imputation procedure can be well used to compensate for missing data, in particular, compared to other imputation procedures. An illustrative example using real data is provided.

Keywords

References

  1. Computational Statistics & Data Analysis v.20 Imputation techniques in regression analysis: Looking closely at their implementation Bello, A. L.
  2. Journal of the American Statistical Association v.89 Missing data, imputation, and the bootstrap Efron, B.
  3. Educational and Psychological Measurement v.58 A comparison of four imputation procedures in a two-variable prediction system Hegamin-Younger, C.;Forsyth, R.
  4. Applied survival analysis Hosmer, D. W.;Lemeshow, S.
  5. Statistical analysis with missing data Little, R. J. A.;Rubin, D. B.