IE SEMINAR: Sparsity Penalized Mean-Variance Portfolio Selection: Analysis and Computation.
Guest: Deniz Akkaya
Title: Sparsity Penalized Mean-Variance Portfolio Selection: Analysis and Computation.
Date / Time: Aprıl, 15, 2026, 13:40
Location: https://sabanciuniv.
Abstract: This work studies mean-variance portfolio selection with an ℓ0-penalty to promote sparsity. The structure of local and global minimizers is characterized and exploited in the development of a branch-and-bound algorithm. Building on this framework, the model is further extended to incorporate uncertainty in the mean return vector via an ellipsoidal uncertainty set. Extensive computational experiments on real data, along with comparisons against an off-the-shelf and a state-of-the-art MIQP solver, are presented.
Bio: Deniz Akkaya is a Ph.D. Candidate and Research Assistant in the Industrial Engineering Department at Bilkent University. His research specializes in sparse optimization, robust regression, and mathematical programming. His work focuses on the theoretical and computational aspects of optimization, particularly analyzing the structure of minimizers for non-differentiable and non-convex functions under sparsity regularizations. He earned his B.Sc. in Mathematics and M.Sc. in Industrial Engineering from Bilkent University.