Identifying Weakly Correlated Dominating Stocks using Maximum Independent Set

Yeremia Bertolla Gimanjar (School of Data Science, Mathematics, and Informatics, IPB University, Bogor, Indonesia)
Diah Prastiwi (Department of Mathematics, Faculty of Military Mathematics and Natural Sciences, Universitas Pertahanan, Bogor, Indonesia)
Fendy Septyanto (School of Data Science, Mathematics, and Informatics, IPB University, Bogor, Indonesia)

Indonesian Actuarial Journal
Vol. 1 No. 2 (2025)

In this paper we consider the problem of efficiently finding a (small) set of stocks taken from an index that can replicate the index performance. Furthermore, we add the requirement that the set’s returns have weak correlation with each other. Such a selection of stocks may be useful for investors who want to simplify their analysis of the stock index, trying to capture market movement with reduced risk. To solve this problem, we use maximum independent set, a concept from graph theory. As a case study we consider IDX80 in the year 2024.