Uniform exponential convergence of sample average random functions under general sampling with applications in stochastic programming (Q973997): Difference between revisions

Given a function depending on a random variable, a well-known method for estimating the related expectation value function is to use the sample mean function. For an increasing sample size, the type and rate of convergence of the sample mean function to the expectation value function is considered under certain assumptions on the class of random functions, as e.g. certain lower semicontinuous random functions. Moreover, the convergence of estimators for stationary points of stochastic programming problems and Nash equilibrium problems based on replacing the occurring expectations by sample means are considered.