Title :
Optimal prediction-intervals for the exponential distribution, based on generalized order statistics
Author :
Raqab, Mohammad Zayed
Author_Institution :
Dept. of Math., Jordan Univ., Amman, Jordan
fDate :
3/1/2001 12:00:00 AM
Abstract :
This paper proposes optimal prediction intervals for the future generalized order statistic (GOS) based on the first r GOS (ordinary order statistics, usual record values, k-record values) from an exponential population. A conditional argument is considered for obtaining an optimal prediction interval for future GOS. A numerical example illustrates this technique
Keywords :
exponential distribution; maximum likelihood estimation; prediction theory; reliability theory; statistical analysis; conditional argument; exponential distribution; generalized order statistics; k-record values; optimal prediction intervals; ordinary order statistics; reliability; usual record values; Distribution functions; Electric shock; Error analysis; Exponential distribution; Maximum likelihood estimation; Predictive models; Probability density function; Random variables; Statistical distributions; Voltage;
Journal_Title :
Reliability, IEEE Transactions on