This project uses advanced optimization algorithms to tackle two key problems:
-
Investment Strategy Optimization:
- Objective: Maximize the Sharpe ratio for investments in three companies.
- Method: Implemented the Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm to iteratively adjust investment allocations and maximize risk-adjusted returns.
-
Queueing Model Optimization:
- Objective: Minimize the average waiting time in a GI/GI/1 queueing system.
- Method: Utilized a Stochastic Approximation (SA) algorithm to optimize service times, ensuring efficient customer service and reduced waiting times.
- Jiakang Huang
- Hongkai Liu
- Shengzhe Ji
- Jiahe Jiang
- Chenghao Cao
- Dr. Bernd Heidergott
- Franssen Christian
- Zehao Li
- Simultaneous Perturbation Stochastic Approximation (SPSA): Efficiently optimized complex, noisy functions by estimating gradients with minimal computational cost.
- Stochastic Approximation (SA): Improved queueing performance through iterative parameter adjustments based on stochastic models.
- Simulation and Statistical Analysis: Conducted extensive simulations to validate algorithm performance, analyze convergence, and ensure robust results.
- Clone the repository.
- Open the Jupyter Notebook.
- Run the code cells to see the simulation results.