"Bath tub shaped" ,  what is that ? It does not qualify anything. Please be more specific.... 

Hi,

Dr Mohanty - I believe "bathtub shaped" failure means a distribution with high early failure, then a relatively long period of low failure risk, then increased frequency of failure again at the high end.

This is similar to general population life expectancy in the absence of famine or war, relatively high mortality rates in the first few years of life through adolescence, then a fairly long floor during young - middle adulthood where mortality risk is low, then increased mortality in the elderly.

Weibul distribution can be used to fit bathtub shaped failure for parametric survival analysis.  I think Klein and Moschberger discuss this (though hardly "latest" research) in their survival analysis text. 

Agree with Dr. Seth

seth@can you provide the full name of the paper by Moschberger

Sure,

its a standard survival analysis text, most of the focus is on standard non/semi parametric approaches (ie Kaplan-Meier / Cox models), but early on there is a good discussion around parametric models and model fitting as I recall. 

Survival Analysis: Techniques for Censored and Truncated Data (Statistics for Biology and Health) Hardcover – March 10, 2005

by John P. Klein (Author), Melvin L. Moeschberger (Author)

ISBN-13: 978-0387953991 ISBN-10: 038795399X Edition: 2nd

Thanks Seth