State-Space Size Estimation By Least-Squares Fitting

Abstract

We present a method for estimating the number of states in the continuous time Markov chains (CTMCs) underlying high-level models using least-squares fitting. Our work improves on existing techniques by producing a numerical estimate of the number of states rather than classifying the state space into on of three types. We demonstrate the practicality and accuracy of our approach on a number of CTMCs generated from three Generalised Stochastic Petri Net (GSPN) models with up to 11 million states.