This paper proposes the sequential procedure which can be used to determine a truncation point to reduce or remove bias owing to artificial startup conditions in simulations aimed at estimating steady-state behavior. It is based on the idea of Lyapuno...
This paper proposes the sequential procedure which can be used to determine a truncation point to reduce or remove bias owing to artificial startup conditions in simulations aimed at estimating steady-state behavior. It is based on the idea of Lyapunov exponent in chaos theory. The performance measures considered are relative bias, coverage, estimated relative half-width of the confidence interval, mean amount of deleted data.