Abstract–Two classes of multiplicative bias correction (MBC) methods with discrete associated kernels are used for probability mass function estimation. The MBC estimators reduce the order of magnitude in bias, whereas the order of magnitude in variance remains unchanged. For the choice of the bandwidth, we proposed the Bayes global method against the unbiased cross-validation approach. The performance of both methods (Bayesian and cross validation methods) are evaluated under two criteria: the integrated square error criteria and the integrated square bias through simulation studies. We also present a real data application of the proposed methods. The obtained results show that the Bayes global method performs better than cross-validation one for MBC estimators.
11 January 2022 in Communications in Statistics - Theory and Methods
In this paper, we demonstrate that the multiplicative bias correction (MBC) approaches can be extended for both Inverse Gamma (IG) and Beta Prime (BP) kernel density estimators. First, some properties of the MBC-IG and MBC-BP kernel density estimators (bias, variance and mean integrated squared error) are shown. Second, the least square cross validation technique (LSCV) is adapted for the choice of bandwidth. Finally, the performances of the MBC estimators based on IG and BP kernels are illustrated by two studies, a simulation study, followed by a real application study with positive support. In general, in terms of the two criterions, integrated squared bias (ISB) and integrated squared error (ISE), the proposed estimators outperform the classical IG and BP kernel estimators.
18 August 2019 in Communications in Statistics - Simulation and Computation
Two multiplicative bias correction (MBC) approaches for nonparametric multivariate associated kernel estimators for joint probability mass functions in the context of discrete supported data are proposed. Both techniques reach an optimal rate of convergence of the mean integrated squared error. We show some properties of the MBC multivariate associated kernel estimators like bias, variance and mean integrated square error. A simulation study and an application on a real count data set illustrate the performance of the MBC estimators based on the Dirac Discrete Uniform associated kernel in terms of the integrated squared error and integrated squared bias.
6 July 2026 in Communications in Statistics: Case Studies, Data Analysis and Applications
1. Estimating a probability density function (pdf) is a fundamental task in statistical analysis. When parametric assumptions do not hold, traditional maximum likelihood estimation (MLE) may become...