wangziy4-3 / adp-portfolio-selection

Approximate Dynamic Programming for Portfolio Selection Problem

ADPPortfolioSelection

Introduction

This is the Python project corresponding to my Master Thesis "Stochastic Dyamic Programming applied to Portfolio Selection problem". My report can be found on my ResearchGate profile. This project is also in the continuity of another project, which is a study of different risk measures of portfolio management, based on Scenarios Generation.

Installation

This project uses Python version 3. However, you have to use python3.5 maximum, because you need to install gurobipy, which is the Python API of an optimisation library called Gurobi.

I strongly encourage you to install the project inside a virtualenv environment:

virtualenv -p python3.5 env
source env/bin/activate

Then, the main dependencies can be installed via pip:

pip install -r requirement.txt

You can download Gurobi on their website and install it. Then, go into the directory (for instance, /Library/gurobi702/mac64 for gurobi v7.02 for Mac 64-bits), and launch (while you're still in the python3.5 virtual environment):

python setup.py install

You should be setup to launch the project!