Estimation of heterogeneous Poisson process parameters using the PSO algorithm compared with conventional methods
AL-Anbar University journal of Economic and Administration Sciences
2019, Volume 11, Issue 26, Pages 578-582
2019, Volume 11, Issue 26, Pages 578-582
Abstract
In this paper, it was proposed to use the PSO algorithm to estimate the rate of occurrence of the Non-homogeneous Poisson process with rate described as a power law function. The results of the estimation method were compared with classical methods of maximum likelihood and least squares. The research included a realistic application in which the operating periods between the two stops per days of the Mosul Gas Station, and estimated the time of the station stop using the estimation methods proposed in the research. Among the applied results, it was observed that the PSO algorithm was better in estimating the rate of occurrence.
Keywords
Non-homogeneous Poisson process, PSO algorithm, power law function, maximum likelihood, least squaresMetrics