Changepoint models are considered for three situations in which the models allow departures from proportional hazards to be approximated in a simple way. The first situation is a so-called time-covariate qualitative interaction (O'Quigley and Pessione 1991) or crossing hazards problem. The second situation considers a decline in regression effect that is modeled by a sudden change in effect at some unknown time point (O'Quigley and Natarajan 2004). The third situation, common in prognostic modeling, deals with inference when we wish to simplify a continuous covariate into two classes, above and below some threshold. The results of Davies (1977, 1987) are particularly useful and the conditions under which these results can be applied are described.
Access provided by Autonomous University of Puebla. Download to read the full chapter text
Chapter PDF
Keywords
- Bootstrap Sample
- Bootstrap Distribution
- Acute Graft Versus Host Disease
- Bootstrap Repetition
- Cytomegalovirus Pneumonia
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Rights and permissions
Copyright information
© 2008 Springer Science+Business Media, LLC
About this chapter
Cite this chapter
(2008). Inference: Changepoint models. In: Proportional Hazards Regression. Statistics for Biology and Health. Springer, New York, NY. https://doi.org/10.1007/978-0-387-68639-4_12
Download citation
DOI: https://doi.org/10.1007/978-0-387-68639-4_12
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-25148-6
Online ISBN: 978-0-387-68639-4
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)