2021/05/14 by Yuriy Krepkiy, Krepkiy, Yuriy, Asif Lakhany +3 · 1 citation
Computer Science · Decision Sciences · Economics, Econometrics and Finance · #Image and Signal Denoising Methods #Probabilistic and Robust Engineering Design #Stochastic processes and financial applications
paper · pdf · doi:10.48550/arxiv.2105.07061
We describe a regression-based method, generally referred to as the Least Squares Monte Carlo (LSMC) method, to speed up exposure calculations of a portfolio. We assume that the portfolio contains several exotic derivatives that are priced using Monte-Carlo on each real world scenario and time step. Such a setting is often referred to as a Monte Carlo over a Monte Carlo or a Nested Monte Carlo method.