Acta Univ. Sapientiae, Economics and Business, 13 (2025) 146–164
Abstract. This study examines the index tracking problem using constrained optimization with sector weight and turnover constraints. We implement a mixed-integer quadratic programming (MIQP) model using the Gurobi framework. Our goal is to construct concentrated portfolios that replicate the Standard & Poor’s (S&P) 100 performance. The approach employs a rolling window methodology with in-sample optimization and out-of-sample validation periods to assess performance across varying market conditions. The analysis evaluates constraint combination effects on tracking accuracy, portfolio stability, and computational requirements for 10- and 20-stock portfolios. Results demonstrate that constraints improve index alignment and portfolio stability, while enabling substantial computational efficiency gains primarily through solution space reduction.
Keywords: index tracking, portfolio optimization, mixed-integer programming, Gurobi solver
SAPIENTIA HUNGARIAN UNIVERSITY OF TRANSYLVANIA
The Sapientia Hungarian University of Transylvania is the independent university of the Hungarian community in Romania, which aims at providing education to the members of our community and performing scientific research on a high professional level.
Sapientia Hungarian University of Transylvania,
Scientia Publishing House
Acta Universitatis Sapientiae
RO 400112 Cluj-Napoca, Romania
Str. Matei Corvin nr. 4.
Email: acta @ acta.sapientia.ro
