Title :
Maximum likelihood estimation of Weibull parameters for two independent competing risk
Author :
Ishioka, Tsunenori ; Nonaka, Yasuo
Author_Institution :
Ricoh Co. Ltd., Tokyo, Japan
fDate :
4/1/1991 12:00:00 AM
Abstract :
A stable technique for obtaining the maximum-likelihood estimate of Weibull parameters of the life distributions of two components that form a series system is presented. The technique requires much more computation than the procedure of H. Kanie and Y. Nonaka (1985). Simulation results, however, show that the standard deviation of the estimated values of the Weibull parameters is greatly reduced. This technique does not require the concomitant indicator and can be applied not only for complete data but for randomly censored data
Keywords :
parameter estimation; probability; reliability theory; Weibull parameters; complete data; maximum likelihood estimation; randomly censored data; reliability theory; Life estimation; Life testing; Maximum likelihood estimation; Newton method; Parameter estimation; Probability; Shape; State estimation; Statistical distributions; Weibull distribution;
Journal_Title :
Reliability, IEEE Transactions on