RANDOM THRESHOLD MODELS APPLIED TO INFLATED ZERO CLASS DATA
Authors:
- 1 March 1997
- journal article
- Published by Wiley in Australian Journal of Statistics
- Vol. 39 (1) , 5-16
Abstract
Summary: For a random component threshold model, the threshold parameters are functions of a linear combination of fixed and random components. This paper describes estimation procedures in terms of composite link functions. The models give a good fit to the frequency of use of medical procedures in the different counties of Washington State, USA. Suitable models for studying change in county effects in successive years are developed.Keywords
This publication has 17 references indexed in Scilit:
- The derivation of blup, ML, REML estimation methods for generalised linear mixed modelsCommunications in Statistics - Theory and Methods, 1995
- Laplace's approximation for nonlinear mixed modelsBiometrika, 1993
- The analysis of proportions in agricultural experiments by a generalized linear mixed modelStatistica Neerlandica, 1993
- Approximate Inference in Generalized Linear Mixed ModelsJournal of the American Statistical Association, 1993
- Can Small-Area Analysis Detect Variation in Surgery Rates?Medical Care, 1992
- Zero-Inflated Poisson Regression, with an Application to Defects in ManufacturingTechnometrics, 1992
- On the Statistical Analysis of Ordinal Data When Extravariation is PresentJournal of the Royal Statistical Society Series C: Applied Statistics, 1990
- Small-Area Variations in the Use of Common Surgical Procedures: An International Comparison of New England, England, and NorwayNew England Journal of Medicine, 1982
- Composite Link Functions in Generalized Linear ModelsJournal of the Royal Statistical Society Series C: Applied Statistics, 1981
- Small Area Variations in Health Care DeliveryScience, 1973