Bayesian Classification with First Order Moving Average Sources
Haroun, A., Shaarawy, S. (1992). Bayesian Classification with First Order Moving Average Sources. EKB Journal Management System, 36(2), 317-325. doi: 10.21608/esju.1992.314869
Ahmed Haroun; Samir Shaarawy. "Bayesian Classification with First Order Moving Average Sources". EKB Journal Management System, 36, 2, 1992, 317-325. doi: 10.21608/esju.1992.314869
Haroun, A., Shaarawy, S. (1992). 'Bayesian Classification with First Order Moving Average Sources', EKB Journal Management System, 36(2), pp. 317-325. doi: 10.21608/esju.1992.314869
Haroun, A., Shaarawy, S. Bayesian Classification with First Order Moving Average Sources. EKB Journal Management System, 1992; 36(2): 317-325. doi: 10.21608/esju.1992.314869

Bayesian Classification with First Order Moving Average Sources

The Egyptian Statistical Journal
Article 10, Volume 36, Issue 2, December 1992, Page 317-325  XML
Document Type: Original Article
DOI: 10.21608/esju.1992.314869
View on SCiNiTO View on SCiNiTO
Authors
Ahmed Haroun; Samir Shaarawy
Abstract
The main objective of this paper is to develop a convenient Bayesian procedure that can be used to assign a univariate time series realization to one of several first order moving average sources, with unknown coefficients, that share a common unknown precision. The foundation of the proposed procedure is to develop the marginal posterior mass function of a classification vector using an approximate conditional likelihood function. A time series realization is assigned to that first order moving average process with the largest posterior probability. A comprehensive simulation study with two sources is carried out to demonstrate the performance of the proposed procedure and to check its adequacy in handling the classification problems.
Keywords
Moving Average Processes; Classification; Posterior Mass Function; Bayesian Analysis
Statistics
Article View: 19