A Method for Comparing Hedge Funds
This work addresses the challenge for investors in identifying similar hedge funds based on time-series data, though it appears incremental as it builds on existing classification techniques with specific enhancements.
The paper tackled the problem of comparing hedge funds by developing machine learning methods for classifying time-series of varying lengths, types, and quantities, and applied them to analyze monthly returns of 11,312 hedge funds from 2000-2010 to identify behavioral similarities and assist investors in finding alternative investments.
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.