Change points detection of a smooth estimate of a hazard function

Warning

This publication doesn't include Faculty of Sports Studies. It includes Faculty of Science. Official publication website can be found on muni.cz.

Authors	HOROVÁ Ivanka POSPÍŠIL Zdeněk ZELINKA Jiří
Year of publication	2010
Type	Conference abstract
MU Faculty or unit	Faculty of Science
Citation
Description	This contribution extends the procedure introduced in J.Theor.Biol. 258(3) 437-443 (2009) combining parametric and non-parametric estimation of hazard function for cancer patients has been developed. The procedure utilizes kernel smoothing of hazard function to identify parameters of deterministic model. The poster presents methods for estimation of roots of the first and the second derivatives of hazard function and compares the estimates with the roots of the identified parametric form of the hazard function. The mentioned roots indicate time of maximal risk and that of the most rapid risk change. The procedure is applied to the breast cancer and to the leukemia data.
Related projects:	Jaroslav Hájek Center for Theoretical and Applied Statistics Diferenční rovnice a dynamické rovnice na time scales III