Multi-Level Monte Carlo with Python (MLMCPy)(LAR-19515-1)

The Multi-Level Monte Carlo with Python (MLMCPy) software package is code written in Python to solve uncertainty propagation problems. Multi-level Monte Carlo (MLMC) is an efficient alternative to standard Monte Carlo simulation for estimating expectations of outputs to computational models with uncertain input parameters. MLMC greatly reduces computational cost by performing most simulations with low accuracy at a correspondingly low cost, with relatively few simulations being performed at high accuracy and high cost. MLMCPy is a parallel, efficient, and modular implementation of the MLMC method that provides a straightforward means of applying the method to general uncertainty propagation problems.