A general framework for statistical inference on discrete event systems.
We present a framework for statistical analysis of discrete event systems which combines tools such as simulation of marked point processes, likelihood methods, kernel density estimation and stochastic approximation to enable statistical analysis of the discrete event system, even if conventional approaches fail due to the mathematical intractability of the model. The approach is illustrated with an application to modelling and estimating corrosion of steel gates in the Dutch Haringvliet storm surge barrier.
|Keywords||discrete event systems, kernel density estimation, likelihood methods, market point process, optimization via simulation, parameter estimation, stochastic approximation|
Nicolai, R.P., & Koning, A.J.. (2006). A general framework for statistical inference on discrete event systems. (No. EI 2006-45). Report / Econometric Institute, Erasmus University Rotterdam. Retrieved from http://hdl.handle.net/1765/8068