Abstract
AbstractWe present a new methodology for modelling the joint distribution of ordered categorical variables with finite mixtures where hypotheses of interest may be expressed by linear equality and inequality constraints on the parameters. The connection with non‐parametric polytomous item response theory models is outlined and an application to the quality of life of asthmatic patients is examined. An algorithm for constrained maximum likelihood estimation is described and an analysis of deviance table for hypotheses testing based on the asymptotic distribution of the likelihood ratio statistic is outlined. Copyright © 2004 John Wiley & Sons, Ltd.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
