Fund2Vec: Mutual Funds Similarity Using Graph Learning

Explore a groundbreaking approach to identifying similar mutual funds and exchange-traded funds using graph learning techniques. Dive into the Fund2Vec method, which leverages a weighted bipartite network representation of funds and their underlying assets. Learn how this sophisticated machine learning approach, based on Node2Vec, creates a low-dimensional embedding of the network to compute node similarities. Discover how this data-driven method provides novel insights into portfolio similarity, removes bias from qualitative categorizations, and offers applications in fund recommender systems, competitor analysis, and product marketing. Gain a deeper understanding of the first-ever study on the weighted bipartite network representation of funds-assets networks, presented by Dhagash Mehta, Senior Manager of ML & Asset Allocation at Vanguard.