Advances and Applications in Statistics
Volume 62, Issue 1, Pages 31 - 53
(May 2020) http://dx.doi.org/10.17654/AS062010031 |
|
USING NON-LINEAR PROGRAMMING TO DETERMINE WEIGHTED COEFFICIENT OF BALANCED LOSS FUNCTION FOR ESTIMATING PARAMETERS AND RELIABILITY FUNCTION OF WEIBULL DISTRIBUTION
Fuad Alduais
|
Abstract: The main goal of this paper is to determine the weighted coefficients of balanced loss function by using non-linear programming instead of determined arbitrary as in literature. The comparison study between maximum likelihood and Bayesian estimator under different loss functions is conducted via Monte Carlo simulation. The evaluation is depended on the values of mean square error. The outputs of the simulation showed that the balanced general entropy loss function has the best performance. Moreover, the simulation verified that the balanced loss function is always better than corresponding loss function. |
Keywords and phrases: balanced loss function, weighted coefficients, Bayesian estimation, non-linear programming.
|
|
Number of Downloads: 255 | Number of Views: 800 |
|