Outcome-driven Feature Discovery using Evolutionary Computation for Sovereign Investment Policy-making
Open Access Depositedidentifying critical performance driver features via symbolic regression and analysis of variance, assessing the performance disparity between individual models and ensembles of models, and contrasting the predictive accuracy of different sets of features created using symbolic regression. The outcome is a decision support framework that enhances SWF policy efficacy via evolved optimal investment decision rules, offering tractable mathematical formulae for policy guidance. This innovative use of evolutionary computation to sovereign wealth fund policy formulation has significant implications for institutional investors and economic policymakers.
This research presents a goal-oriented discovery framework that integrates symbolic regression modeling with target policy outcomes to improve investment strategy formulation for sovereign investors, particularly sovereign wealth funds (SWFs). The study examines a critical challenge where sovereigns struggle to design models that generate desired macroeconomic outcomes, leading to a decline in policy efficacy and constraining investment objective fulfillment. The modeling framework utilizes licensed private equity datasets from 2021 to 2024, public equities datasets from 2010 to 2020, and bond yields data from 1961 to 2024, using genetic programming as the engine to develop ensembles of mathematically evolved models. The methodology centers on three primary research inquiries
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.
| Thumbnail | Title | Date Uploaded | Visibility | Actions |
|---|---|---|---|---|
|
|
VenegasObando_gwu_0075A_17699.pdf | 2025-12-12 | Open Access |
|