Dakota Reference Manual
Version 6.4
LargeScale Engineering Optimization and Uncertainty Analysis

Stopping criterion based on relative error
Alias: none
Argument(s): REAL
Default: 1.e4
Multilevel sampling seeks an error balance between the estimator variance and the remaining bias error at the highest level, the two contributors to mean squared error (MSE). Since the remaining bias error is generally unknown, the convergence_tolerance is used to provide a error target relative to the Monte Carlo estimator variance resulting from the pilot sample. The samples allocated at each level are proportional to , so each order of magnitude reduction in convergence_tolerance will tend to increase the sample allocation by two orders of magnitude. Therefore, this control should be used with care to avoid allocation of huge sample sets that could overrun available memory.
Default Behavior
The default value for convergence_tolerance is currently .0001, which may be too resolved for expensive simulations or high variance QoI.