Pre-integration via Active Subspaces

TL;DR

This paper introduces an active subspace-based pre-integration method for quasi-Monte Carlo that improves variance reduction and accuracy in high-dimensional Gaussian integrals, with theoretical and numerical validation.

Contribution

It proposes using the first eigenvector of an active subspace for pre-integration, demonstrating competitive performance and theoretical variance reduction guarantees.

Findings

Active subspace pre-integration is effective for Asian options.

Outperforms traditional methods on basket options.

Theoretically reduces variance with scrambled net integration.

Abstract

Pre-integration is an extension of conditional Monte Carlo to quasi-Monte Carlo and randomized quasi-Monte Carlo. It can reduce but not increase the variance in Monte Carlo. For quasi-Monte Carlo it can bring about improved regularity of the integrand with potentially greatly improved accuracy. Pre-integration is ordinarily done by integrating out one of $d$ input variables to a function. In the common case of a Gaussian integral one can also pre-integrate over any linear combination of variables. We propose to do that and we choose the first eigenvector in an active subspace decomposition to be the pre-integrated linear combination. We find in numerical examples that this active subspace pre-integration strategy is competitive with pre-integrating the first variable in the principal components construction on the Asian option where principal components are known to be very effective. It outperforms other pre-integration methods on some basket options where there is no well established default. We show theoretically that, just as in Monte Carlo, pre-integration can reduce but not increase the variance when one uses scrambled net integration. We show that the lead eigenvector in an active subspace decomposition is closely related to the vector that maximizes a less computationally tractable criterion using a Sobol' index to find the most important linear combination of Gaussian variables. They optimize similar expectations involving the gradient. We show that the Sobol' index criterion for the leading eigenvector is invariant to the way that one chooses the remaining $d-1$ eigenvectors with which to sample the Gaussian vector.