The Success of AdaBoost and Its Application in Portfolio Management

TL;DR

This paper explains why AdaBoost is effective, introduces a measure of noise influence, and demonstrates its application in portfolio management with empirical validation in the Chinese market.

Contribution

It introduces the influence of noise points (ION) as a new measure and links it to AdaBoost's success, providing theoretical insights and empirical evidence.

Findings

ION decreases with more iterations and complex base learners

Deep trees are necessary for consistent classification in complex cases

Empirical studies in Chinese market support theoretical claims

Abstract

We develop a novel approach to explain why AdaBoost is a successful classifier. By introducing a measure of the influence of the noise points (ION) in the training data for the binary classification problem, we prove that there is a strong connection between the ION and the test error. We further identify that the ION of AdaBoost decreases as the iteration number or the complexity of the base learners increases. We confirm that it is impossible to obtain a consistent classifier without deep trees as the base learners of AdaBoost in some complicated situations. We apply AdaBoost in portfolio management via empirical studies in the Chinese market, which corroborates our theoretical propositions.