Online Learning of Commission Avoidant Portfolio Ensembles

TL;DR

This paper introduces an online ensemble learning method for portfolio selection that effectively manages transaction costs through a novel commission avoidance mechanism, achieving low regret and outperforming existing approaches.

Contribution

It presents a new online ensemble strategy that controls and exploits multiple portfolio algorithms while minimizing transaction costs and providing theoretical regret guarantees.

Findings

Proves logarithmic regret bound for the proposed strategy.

Demonstrates significant empirical improvement over state-of-the-art methods.

Validates the approach with numerical experiments showing practical viability.

Abstract

We present a novel online ensemble learning strategy for portfolio selection. The new strategy controls and exploits any set of commission-oblivious portfolio selection algorithms. The strategy handles transaction costs using a novel commission avoidance mechanism. We prove a logarithmic regret bound for our strategy with respect to optimal mixtures of the base algorithms. Numerical examples validate the viability of our method and show significant improvement over the state-of-the-art.