FIDUCIAL DISTRIBUTION TO ESTIMATE THE PARAMETERS OF LOGNORMAL DISTRIBUTION FOR COMPLETE, CENSORED OR ZERO INFLATED DATA
Abstract: Fiducial inference was introduced in the first half of the 20th century by Fisher (1935). While the method originally fell out of favor due to nonexactness issues in multivariate cases, the method has garnered renewed interest in the last decade. Fiducial inference attempts to find a middle ground between the frequentist and Bayesian perspectives. The fiducial argument allows one to fit a posterior-like distribution on a target parameter θ in a way that is entirely data-driven and does not rely on a sometimes-arbitrary prior selection. While the strengths and limitations of generalized fiducial inference (GFI) continue to be explored, its usefulness has already been illustrated in numerous practical applications. This talk will be focusing on implementation of fiducial distributions to estimate different parameters of lognormal distribution based on complete, censored or zero-inflated data. Practical examples will demonstrate that the fiducial methods are accurate and practically exact even for small samples.
This talk is appropriate for faculty, undergraduate students, and a more general audience.
Biography: Md Sazib Hasan is an Assistant Professor of Mathematics in the Department of Mathematics at Utah Tech University. He has been teaching at UT since 2019. He received his PhD. in Statistics from University of Louisiana at Lafayette in 2019. His doctoral dissertation was in fiducial inference for Lognormal distribution which primarily applied to real-world issues in environmental sciences and pollution studies.
His research interest lies in small sample problems, statistical inference, and statistical learning. Hasan is a Project-NExT fellow and demonstrates a keen interest in undergraduate research, evidenced by his mentorship resulting in twenty-four student presentations across local, regional, and national conferences. Currently, he is serving as a First Vice Chair at the MAA Intermountain Section.
Contact: MdSazib.Hasan@utahtech.edu