Kernel Estimation of Conditional Hazard Function for  Cancer Data

Warning

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

Authors	SELINGEROVÁ Iveta HOROVÁ Ivanka ZELINKA Jiří
Year of publication	2014
Type	Article in Proceedings
Conference	Recent Advances in Energy, Environment, Biology and Ecology
MU Faculty or unit	Faculty of Science
Citation
Web	http://www.google.cz/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&sqi=2&ved=0CDQQFjAB&url=http%3A%2F%2Fwww.wseas.us%2Fbooks%2F2010%2FCambridge%2FEE.pdf&ei=-MbXUoS3AefgygPPvYDABA&usg=AFQjCNHWU0JL2ynkTAX-FLuET2h9wLM7rg&sig2=w7NY0ORsFF3I-zO1oBC1uQ&bvm=b
Field	Applied statistics, operation research
Keywords	Hazard function; kernel; bandwidth; cross-validation method; survival function; censoring
Description	The hazard function is a useful tool in survival analysis and reflects the instantaneous probability that an individual will die within the next time instant. In practice, the hazard function depends on covariates as an age and a gender. The most frequently used method to estimate a conditional hazard function is semiparametric model suggested by D. R. Cox. Assumptions of this model are too restrictive in many cases. In the present paper is proposed an estimator for conditional hazard function as the ratio of kernel estimators for the onditional density and survival function. We illustrate the utility of the proposed method through application to cancer data sets.