| Full text | |
| Author(s): |
Al-Sharadqah, Ali A.
;
Patriota, Alexandre G.
Total Authors: 2
|
| Document type: | Journal article |
| Source: | JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION; v. 92, n. 10, p. 29-pg., 2022-02-04. |
| Abstract | |
This paper presents the method-of-moments (MM) and the maximum likelihood (ML) estimators for the width of a uniform random variable U when measured with additive errors. We study situations where the support is either the symmetric interval [-a, a] or asymmetric interval [a, b]. Both MM and ML estimators for the boundaries are discussed in two cases, namely, when the noise level sigma(2) is known and when it is unknown. While the MM estimators have a closed form, the ML estimators need an iterative algorithm to solve a system of nonlinear equations. A reliable algorithm to compute the ML estimators is proposed. We also establish sufficient conditions for the existence of the MM estimators and their asymptotic normality. Numerical experiments on real and synthetic data are presented to validate our results. (AU) | |
| FAPESP's process: | 18/21934-5 - Network statistics: theory, methods, and applications |
| Grantee: | André Fujita |
| Support Opportunities: | Research Projects - Thematic Grants |