Bias-corrected maximum likelihood estimation for the beta distribution
- 1 May 1997
- journal article
- research article
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 58 (1) , 21-35
- https://doi.org/10.1080/00949659708811820
Abstract
This paper gives closed-form expressions for bias-corrected maximum likelihood estimates of the parameters of the beta distribution that can be used to define bias-corrected estimates that are nearly unbiased. Some approximations based on asymptotic expansions for the bias corrections are given. We also present simulation results comparing the performances of the maximum likelihood estimates and corrected ones. The results suggest that bias-corrected estimates have better finite-sample performance than standard maximum likelihood estimates.Keywords
This publication has 18 references indexed in Scilit:
- Bias correction in ARMA modelsStatistics & Probability Letters, 1994
- Bias reduction of maximum likelihood estimatesBiometrika, 1993
- Parameter estimation for the beta distributionJournal of Statistical Computation and Simulation, 1992
- Bartlett corrections and bias correction for two heteroscedastic regression modelsCommunications in Statistics - Theory and Methods, 1992
- Effective procedures for estimating beta distribution's parameters and Their confidence intervalsJournal of Statistical Computation and Simulation, 1991
- On parameter transformations and interval estimationBiometrika, 1984
- A Note on Bias Correction in Maximum Likelihood Estimation with Logistic DiscriminationTechnometrics, 1980
- Maximum likelihood estimation for the beta distributionJournal of Statistical Computation and Simulation, 1978
- Algorithm AS 103: Psi (Digamma) FunctionJournal of the Royal Statistical Society Series C: Applied Statistics, 1976
- Maximum Likelihood Estimation of the Parameters of the Beta Distribution from Smallest Order StatisticsTechnometrics, 1967