Advances and Applications in Statistics
Volume 62, Issue 2, Pages 127 - 137
(June 2020) http://dx.doi.org/10.17654/AS062020127 |
|
PERCENTILES ESTIMATION FOR THE PARAMETERS OF POWER FUNCTION DISTRIBUTION
Riffat Jabeen and Azam Zaka
|
Abstract: This research paper deals with estimation of the parameters from power function distribution through the use of symmetrically located percentiles from a sample. We have used simulation to observe the sampling behavior of the estimates. We have used three alternative methods such as method of percentile, method of ridge regression and method of least square and made its comparison with method of maximum likelihood and method of least squares estimators. We have observed that the use of percentile estimator (i.e. the 5th and 95th percentile) is better than method of maximum likelihood estimator in variability and also from least square estimator in accuracy and variability. |
Keywords and phrases: parameter estimation, power function distribution, percentiles, least squares method, ridge regression.
|
|
Number of Downloads: 247 | Number of Views: 717 |
|