A modified maximum likelihood estimation for the three parameters in lognormal distribution model / Faiz Zulkifli, Noorizam Daud and Norazan Mohamed Ramli

Zulkifli, Faiz and Daud, Noorizam and Mohamed Ramli, Norazan (2011) A modified maximum likelihood estimation for the three parameters in lognormal distribution model / Faiz Zulkifli, Noorizam Daud and Norazan Mohamed Ramli. Research Reports. Research Management Institute (RMI), Shah Alam. (Unpublished)

Abstract

The introduction of the threshold parameters in three parameters lognormal distribution(λ ,μ ,σ) creates complications when we seek to estimate these parameters from sample. Hill(1963) has shown that global maximum likelihood estimators resulted in inadmissibleestimates as the likelihood function of any ordered sample tends to infinity when(λ ,μ ,σ ) approach( ,−∞, ∞) 1 x respectively. Hence, in this project we would like to propose anew modified version of maximum likelihood estimation to cater for the above problem. Theperformance of the proposed method compared to the existing method suggested by Cohenand Whitten (1980), will be examined and verified through a rigorous simulation procedureusing S-PLUS programming language. A sensitivity analysis will be conducted to study thebehaviour of the estimators in meeting the asymptotic normality assumption. For illustration,the proposed method will be applied to real data sets such as biological and physical sciencesdata.

Download

[img] Text
LP_FAIZ ZULKIFLI RMI 11_5.pdf

Download (130kB)

Metadata

Item Type: Monograph (Research Reports)
Creators:
CreatorsID Num.
Zulkifli, FaizUNSPECIFIED
Daud, NoorizamUNSPECIFIED
Mohamed Ramli, NorazanUNSPECIFIED
Subjects: Q Science > QA Mathematics > Factor analysis. Principal components analysis. Correspondence analysis
Q Science > QA Mathematics > Instruments and machines
Divisions: Research Management Institute (RMI)
Item ID: 26327
Uncontrolled Keywords: parameters, infinity, S-PLUS
URI: http://ir.uitm.edu.my/id/eprint/26327

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year