A Method for Comparing Hedge Funds

TL;DR

This paper introduces novel machine learning techniques for classifying and analyzing hedge fund time-series data, enabling investors to identify behavioral similarities and alternative investments across categories and locations.

Contribution

It presents new methods—signal composition and self-labeling—for classifying time-series data regardless of length, type, or quantity, applied to hedge fund returns.

Findings

Effective classification of hedge fund time-series

Identification of behavioral similarities among funds

Supports investor decision-making with cross-category analysis

Abstract

The paper presents new machine learning methods: signal composition, which classifies time-series regardless of length, type, and quantity; and self-labeling, a supervised-learning enhancement. The paper describes further the implementation of the methods on a financial search engine system to identify behavioral similarities among time-series representing monthly returns of 11,312 hedge funds operated during approximately one decade (2000 - 2010). The presented approach of cross-category and cross-location classification assists the investor to identify alternative investments.