Graphical representation of two mixed-Weibull distributions

A variety of shapes of two-Weibull mixtures on Weibull probability paper are explored and classified into six types of Cdf curves (A-F). Types B-D represent Cdf's composed of two well-mixed subpopulations. Type F represents the Cdf composed of two very well-separated subpopulations. Types A and...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on reliability 1992-06, Vol.41 (2), p.241-247
Main Authors: Jiang, S., Kececioglu, D.
Format: Article
Language:English
Subjects:
Data analysis
Distribution theory
Exact sciences and technology
Mathematics
Parameter estimation
Probability
Probability and statistics
Sciences and techniques of general use
Shape
Statistical analysis
Statistical distributions
Statistics
Weibull distribution
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A variety of shapes of two-Weibull mixtures on Weibull probability paper are explored and classified into six types of Cdf curves (A-F). Types B-D represent Cdf's composed of two well-mixed subpopulations. Type F represents the Cdf composed of two very well-separated subpopulations. Types A and E are in between types B-D and type F. The Kao-Cran graphical parameter estimation method cannot be applied to type A-C, E, and F curves. It is recommended that it not be applied to the type D curve. The theoretical basis is developed for the Jensen-Petersen graphical method for mixtures with well-separated subpopulations, viz, the type F curve. For type A and E curves, the method can be applied. However, for type B-D curves, there is no theoretical basis for the method.< >
ISSN:0018-9529
1558-1721
DOI:10.1109/24.257789