Skip to content

Jacoposigno1999/Portfolio-optimization-with-DDPG

Repository files navigation

Portfolio-optimization-with-DDPG

In this repo is stored my master thesis. The primary objective of this thesis is to investigate the application of Deep Reinforcement Learning algorithms, to the problem of asset portfolio management. Specifically, we aim to develop and evaluate a DDPG-based agent capable of autonomously managing an investment portfolio in a financial market environment.

Note:

  • The 2 codes differ only in the way the reward is calculated (in Baseline_reward is calculated using the SP500 as benchmark in Relative_reward is calculated using as benchmark a portfolio containing the same stocks as the managed portfolio but keeping the weights constant
  • Code should be better organized into different python files

About

Deep Reinforcement Learning for Autonomous Portfolio Management: A Policy Gradient Approach

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors