Prediction in Random Coefficient Regression Models
β Scribed by Dr. Jan Bondeson
- Publisher
- John Wiley and Sons
- Year
- 2007
- Tongue
- English
- Weight
- 781 KB
- Volume
- 32
- Category
- Article
- ISSN
- 0323-3847
No coin nor oath required. For personal study only.
β¦ Synopsis
Much attention has been given to the problem ofpredicting future obeervatiomfor some individual within a random coefficient regreasion (RCR) model, using the previous observations on that individual aa well es the information from the re& of the data material. In thia paper, the literature on this subject ie critically reviewed and new methode of linear prediction are proposed for the g e n d RCR model. Exact reaulte am derived for the mean squared errom of Bome predictors in a spacial case, but this ie not poeaible in the general RCR model when ite parametem are not known. In this model, the old and new predictom are compared in a simulation efudy, and further illustrated by prediction in a medical data material.
π SIMILAR VOLUMES
RCR models are reviewed. Various variance estimators are described, among them a new one. Thew variance eathatore are compared in a simulation study. An obahtric data aet is subjected to a detailed analysis by meana of RCR techniques. In particular, interval estimation ia considered.
Using an applications perspective *Thermodynamic Models for Industrial Applications* provides a unified framework for the development of various thermodynamic models, ranging from the classical models to some of the most advanced ones. Among these are the Cubic Plus Association Equation of State (CP