On Usefulness of Maximum Likelihood Estimator Using Incomplete Data

  • Hironori Fujisawa
Conference paper


We often encounter with missing data which consist of a complete part and an incomplete part. In this case, there are two types of maximum likelihood estimator. One is the conventional maximum likelihood estimator based on all of the data and another is based on the complete part only, which neglects the incomplete part. Let n and n* be the sample sizes of the complete part and the incompleste part, respectively. It is well-known that the former is asymptotically better than the latter when n*n is constant. However, other cases have not been well-known. This paper shows that the former is better than the latter in other cases in view of higher-order. In addition, this paper illustrates some exact comparison.


Probability Density Function Maximum Likelihood Estimator Likelihood Estimator Exponential Family Fisher Information Matrix 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Amari, S. (1985). Differential-Geometrical Methods in Statistics. Springer-Verlag, New York.CrossRefMATHGoogle Scholar
  2. Fujisawa, H. (2000). On usefulness of incomplete data for maximum likelihood estimator. Hiroshima University Technical Report, 99–19.Google Scholar
  3. Kano, Y., Bentler, P.M., and Mooijaart, A. (1993). Additional information and precision of estimators in multivariate structural models. In Stat.Sci.and Data Anal., Matsusita, K. et al. (eds.), 187–196, VSP, Utrecht.Google Scholar
  4. Kariya, T., Krishnaiah, P.R., and Rao, C.R. (1983). Inference on parameters of multivariate normal populations when some data is missing. In Developments in Statistics, Vol.4“, Krishnaiah, P.R. (eds.), 137–184, Academic Press, New York.Google Scholar
  5. Komaki, F. (1996). On asymptotic properties of predictive distributions. Biometrika, 83, 299–313.MathSciNetCrossRefMATHGoogle Scholar
  6. Konishi, S. and Shimizu, K. (1994). Maximum likelihood estimation of an intraclass correlation in a bivariate normal distribution with missing observations. Commun. Statist. Theory Meth., 23, 1593–1604.MathSciNetCrossRefMATHGoogle Scholar
  7. Kullback, S. (1959). Information Theory and Statistics. Wiley, New YorkMATHGoogle Scholar
  8. Little, R.J.A. and Rubin, D.B. (1987). Statistical Analysis with Missing Data. Wiley, New YorkMATHGoogle Scholar
  9. Miyakawa, M. (1984). Analysis of incomplete data in competing risks model. IEEE Trans. Reliab., 33, 293–296.CrossRefGoogle Scholar
  10. Yanagimoto, T. (1991). Estimating a model through the conditional MLE. Ann. Inst. Statist. Math., 43, 735–746.MathSciNetCrossRefMATHGoogle Scholar

Copyright information

© Springer Japan 2002

Authors and Affiliations

  • Hironori Fujisawa
    • 1
  1. 1.The Institute of Statistical MathematicsMinato-ku TokyoJapan

Personalised recommendations