The total hazard construction, antithetic variates and simulation of stochastic systems

This paper combines recent developments in generation of dependent random variables with the use of common and antithetic random numbers. This combination yields new efficient methods for estimating complicated stochastic quantities by simulation. Some theoretical and practical aspects of antithetic and common random numbers for variance reduction while using the total hazard construction are given. Their optimality in estimating the expected value of the response sum or the response difference of functions of vector arguments with dependent components is proved. Some numerical examples illustrate the theoretical results.