Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/77117 
Erscheinungsjahr: 
2002
Schriftenreihe/Nr.: 
Technical Report No. 2002,66
Verlag: 
Universität Dortmund, Sonderforschungsbereich 475 - Komplexitätsreduktion in Multivariaten Datenstrukturen, Dortmund
Zusammenfassung: 
The correlated Weibull regression model for the analysis of correlated binary data is presented. This regression model is based on Bonney’s disposition model for the regression analysis of correlated binary outcomes. Parameter estimation was done through the maximum likelihood method. The correlated Weibull regression model was contrasted with the correlated logistic regression model. The results showed that both regression models were useful in explaining the familial aggregation of oesophageal cancer. The correlated logistic regression model fitted the oesophageal cancer data better than the correlated Weibull regression model for both the non-nested and nested cases. Furthermore, the correlated logistic regression model was computationally more attractive than the correlated Weibull regression model.
Schlagwörter: 
Correlated binary data
Non-nested disposition model
Nested disposition model
Weibull distribution
Dokumentart: 
Working Paper

Datei(en):
Datei
Größe
402.71 kB
167.14 kB





Publikationen in EconStor sind urheberrechtlich geschützt.