Ensemble learning for portfolio valuation and risk management

TL;DR

This paper presents a fast, accurate ensemble learning approach using regression trees for dynamic portfolio valuation and risk management, capable of handling Bermudan options and scaling efficiently with data size.

Contribution

It introduces a novel ensemble learning method that provides a closed-form estimator for portfolio valuation, improving speed and scalability over existing techniques.

Findings

Method is fast and accurate in moderate dimensions

Scales well with sample size and path space dimension

Effective for Bermudan style options

Abstract

We introduce an ensemble learning method for dynamic portfolio valuation and risk management building on regression trees. We learn the dynamic value process of a derivative portfolio from a finite sample of its cumulative cash flow. The estimator is given in closed form. The method is fast and accurate, and scales well with sample size and path space dimension. The method can also be applied to Bermudan style options. Numerical experiments show good results in moderate dimension problems.