Skip to content

Portfolio optimization in AMPL; compared minimum covariance and Young's minimax models

Notifications You must be signed in to change notification settings

ekapope/Portfolio-Optimization

 
 

Repository files navigation

Portfolio optimization

by: Ekapope Viriyakovithya , James Ponce, Deborah Kewon, Saeed Mirzaee, Usman Muhammad

Introduction

In this study, we aimed to create a model using AMPL to propose investors to effectively manage their stocks and to invest on the most profitable stocks. We collected stock market data of 500 companies from IEX Group and allocated the assets based on the principle of portfolio optimization.

Data description

Firstly, we randomly retrieved 500 stocks from the total 8,718 stocks in the U.S. market with the period of 5 years (1258 trading days). Each row is the closing price of a stock for the designated trading date. However, using student version of AMPL software, we are allowed to compute only up to 300 variables for a non-linear problem (500 for linear problem). Therefore, we omitted 200 stocks in the second step. We aggregated stock prices by mean and grouped by month, and then calculated average percentage changes, covariance, and used them as inputs for AMPL.

For more detail and findings, please refer to Portfolio_Optimization_Group_Project.pdf.

About

Portfolio optimization in AMPL; compared minimum covariance and Young's minimax models

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • R 65.0%
  • Python 25.2%
  • AMPL 9.8%