Abstract
Estimators based on the mode are introduced and shown empirically to have smaller Kullback–Leibler risk than the maximum likelihood estimator. For one of these, the midpoint modal estimator (MME), we prove the Kullback–Leibler risk is below \({\frac{1}{2}}\) while for the MLE the risk is above \({\frac{1}{2}}\) for a wide range of success probabilities that approaches the unit interval as the sample size grows to infinity. The MME is related to the mean of Fisher’s Fiducial estimator and to the rule of succession for Jefferey’s noninformative prior.
Similar content being viewed by others
References
Agresti A. (2002) Categorical data analysis. Wiley, New York
Copas J. (1997) Using regression models for prediction: Shrinkage and regression to the mean. Statistical Methods in Medical Research 6: 167–183
Eguchi S., Yanagimoto T. (2008) Asymptotic improvement of maximum likelihood estimators on Kullback–Leibler loss. Journal of Statistical Planning and Inference 138: 3502–3511
Fisher R. (1973) Statistical methods and scientific inference (3rd ed). Hafner Press, New York
Heskes T. (1998) Bias/variance decompositions for likelihood-based estimators. Neural Computation 10: 1425–1433. doi:10.1162/089976698300017232
Kempthorne O., Folks L. (1971) Probability, statistics, and data analysis. The Iowa State University Press, Ames, IA
Moore D., McCabe G., Craig B. (2009) Introduction to the practice of statistics. Freeman, New York
Vos P., Hudson S. (2005) Evaluation criteria for discrete confidence intervals: Beyond coverage and length. The American Statistician 59: 137–142
Vos P., Hudson S. (2008) Problems with binomial two-sided tests and their associated intervals. Australian & New Zealand Journal of Statistics 50: 81–89
Wu, Q., Vos, P. (2009). Improving boundaries of binomial 50 percentage points with applications to the midpoint modal estimator, East Carolina University. Biostatistics Technical Report #09.2.
Author information
Authors and Affiliations
Corresponding author
About this article
Cite this article
Vos, P., Wu, Q. Estimators for the binomial distribution that dominate the MLE in terms of Kullback–Leibler risk. Ann Inst Stat Math 64, 359–371 (2012). https://doi.org/10.1007/s10463-010-0316-3
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10463-010-0316-3