Estimation of parameters of the continuous uniform distribution: Different classical methods. Issue 3 (2nd April 2020)

• Record Type: Journal Article
• Authors: Jabeen, Riffat; Zaka, Azam
• Is Part Of: Journal of statistics & management systems. Volume 23:Issue 3(2020)
• Page Start: 529
• Page End: 547
• Publication Date: 2020-04-02
• Subjects: 60E05; Maximum likelihood -- Moment estimation -- Total deviation -- Mean square error; Statistics -- Periodicals; Mathematical models -- Periodicals; Mathematical models; Statistics; Periodicals; 519.5
• DOI: 10.1080/09720510.2019.1639948
• Languages: English
• ISSNs: 0972-0510

Abstract

Abstract: The paper is concerned with the continuous uniform distribution for the estimation of its parameters. We have defined the different methods that may be utilized for the estimation of the parameters of a continuous uniform distribution which includes least square and relative least square methods also method of ridge regression, method of moments and its modifications, method of Percentile and its modifications, method of maximum likelihood estimation. The parameters of the continuous uniform distribution are estimated using thirteen different of methods for Monte Carlo generated samples. A repeated simulation study compares these estimation methods. We use the total deviation and mean square as performance index to identify the best estimator amongst them.From the simulation results, we found that the third modified percentile estimator (using 5 th and 95 th percentiles) performs better for the estimation of parameters from the continuous uniform distribution.