EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3595. Dynamic Investment Model for Pension Funds: Maximizing Mean-Risk Performance with SD constraints
Invited abstract in session MD-51: Portfolio risk management, stream Risk management in finance.
Monday, 14:30-16:00Room: M5 (building: 101)
Authors (first author is the speaker)
| 1. | Audrius Kabasinskas
|
| Department of Mathematical Modeling, Kaunas University of Technology | |
| 2. | Milos Kopa
|
| Department of Probability and Mathematical Statistics, Charles University in Prague, Faculty of Mathematics and Physics | |
| 3. | Kristina Sutiene
|
| Department of Mathematical Modeling, Kaunas University of Technology | |
| 4. | Sebastiano Vitali
|
| Department of Economics, University of Bergamo |
Abstract
This study introduces a dynamic investment model for pension fund managers, aiming to optimize mean-risk performance over a 21-year horizon. The model reacts to past asset returns, with annual rebalancing, and seeks to outperform benchmarks using second-order stochastic dominance. We explore the possibility of identifying a universally preferred dynamic strategy for all risk-averse investors and examine its characteristics compared to benchmarks. The model maximizes mean-0.5*CVaR with stochastic dominance constraints, utilizing historical extraction for future returns. Benchmarks follow a 1/N portfolio strategy, evolving asset allocations over time according to pension fund regulations. Results for the benchmark show an average terminal wealth of 2.7 EUR after 21 years, with a 5% probability of falling below 1 EUR. The optimal strategy, without ESG restrictions, outperforms the benchmark, exhibiting higher cash and bond investments, resulting in an expected terminal wealth of 6 EUR. When considering only ESG attractive assets, the optimal strategy is less profitable (expected terminal wealth of 3.6 EUR), yet it outperforms the benchmark in second-order stochastic dominance. The optimal strategy demonstrates sensitivity to asset bounds and ESG considerations, emphasizing the importance of strategic asset selection over risk aversion parameters.
The project has received funding from the Research Council of Lithuania (LMTLT), agreement No S-MIP-21-32.
Keywords
- Multi-Objective Decision Making
- Financial Modelling
- Risk Analysis and Management
Status: accepted
Back to the list of papers