Abstract
Having been observed as one of the most useful distributions for modelling and analyzing lifetime data in medical, biological, engineering and others, the Weibull distribution has been studied enthusiastically in the literature to determine the best method in estimating its parameters. The objective of our study is to determine the reliable estimator among three methods for estimating the parameters of Weibull distribution and how bias the estimators' estimates of the parametersare to the true values. The methods being examined here are maximum likelihood estimator, least square estimators of Y on X and X on Y. The methods are compared using MSE and Bias base on simulation study. From the study it is observed that least square on X is more reliable for estimating the shape parameter with relatively small samples but with larger samples LSY will be the preferred method whiles maximum likelihood is reliable for estimating the scale parameter.
Original language | English |
---|---|
Pages (from-to) | 1329-1334 |
Number of pages | 6 |
Journal | ARPN Journal of Engineering and Applied Sciences |
Volume | 7 |
Issue number | 10 |
Publication status | Published - Oct 2012 |
Externally published | Yes |
Keywords
- Least square estimators
- Maximum likelihood estimator
- Partly censored data
- Simulation study
- Weibull distribution