Parameter estimation for Burr type XII distribution with differential evolution and quasi-Newton approaches based on progressively Type I interval-censored samples

機構典藏 > College of Business and Management > Graduate Institute & Department of Statistics > Journal Article > Item 987654321/113393

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/113393

Title:	Parameter estimation for Burr type XII distribution with differential evolution and quasi-Newton approaches based on progressively Type I interval-censored samples
Authors:	J Zhu;H Xin;J Sun;Y-Y Fan;C Zheng;T-R Tsai
Date:	2017-09
Issue Date:	2018-05-10 12:10:39 (UTC+8)
Abstract:	Burr type XII distribution (BXIID) has earned more attention in the fewpast decades because of the flexibility of distribution shape for model fitting. However,no analytical closed formula solutions of the maximum likelihood estimates (MLEs) ofBXIID parameters can be obtained based on progressively type I interval-censored (PTIIC)samples. In this manuscript, the differential evolution algorithm (DE) and quasi-Newtonmethod (QN) are applied to searching the MLEs of BXIID parameters based on PTIICsamples. The performance of DE and QN is evaluated by means of Monte Carlo simu-lations. Simulation results show that the DE outperforms QN in terms of smaller biasand mean squared error (MSE) of the MLEs.
Relation:	ICIC Express Letters B 8(9), p.1299-1306
DOI:	10.24507/icicelb.08.09.1299
Appears in Collections:	[Graduate Institute & Department of Statistics] Journal Article

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	100	View/Open
Parameter estimation for Burr type XII distribution with differential evolution and quasi-Newton approaches based on progressively Type I interval-censored samples.pdf		106Kb	Adobe PDF	0	View/Open