Remarks on stochastic automatic adjoint differentiation and financial models calibration

01/14/2019

∙

by   Dmitri Goloubentcev, et al.

∙

0

∙

share

In this work, we discuss the Automatic Adjoint Differentiation (AAD) for functions of the form G=1/2∑_1^m (Ey_i-C_i)^2, which often appear in the calibration of financial models. This helps to understand the algorithm proposed recently in Friez by C. Fries. We suggest to use the term Stochastic AAD in situations when expectation is an internal operation. We analyze this in detail and provide the cost estimate of the SAAD for the case when the AAD tool allows an automatic parallelization.