Research partially supported by NSF grant DMS88–03271 and ARO grant DAAL03–91-G-0111. The author would like to express sincere thanks to Professor Wei-Yin Loh for his discussion of this paper and most valuable suggestions.
Tree-Structured Exponential Regression Modeling†
Version of Record online: 18 JAN 2007
Copyright © 1994 WILEY-VCH Verlag GmbH & Co. KGaA
Volume 36, Issue 1, pages 43–61, 1994
How to Cite
Ahn, H. (1994), Tree-Structured Exponential Regression Modeling. Biom. J., 36: 43–61. doi: 10.1002/bimj.4710360106
- Issue online: 18 JAN 2007
- Version of Record online: 18 JAN 2007
- Manuscript Received: AUG 1992
- Exponential regression;
- Recursive partitioning;
- Survival analysis
A method for fitting piecewise exponential regression models to censored survival data is described. Stratification is performed recursively, using a combination of statistical tests and residual analysis. The splitting criterion employed in cross-validation is the average squared error of the residuals. The bootstrap is employed to keep the probability of a type I error (the error of discovering two or more strata when there is only one) of the method close to a predetermined value. The proposed method can thus also serve as a formal goodness-of-fit test for the exponential regression model. Real and simulated data are used for illustration.