Perfect Simulation for a Stationary Silo with Absorbing Walls

The objective of this work is to generate random samples of the unique stationary distribution associated to the stochastic model for grain storage in a finite bidimensional silo. The support of this measure is an unbounded and continuous state space and therefore a truncation was necessary to apply the CFTP perfect simulation scheme. The performance of the algorithm was measured by comparing the sample moments to the theoretical ones.