Forecasting Returns with a Recurrent Neural Network

Project Definition

With the development of deep learning and the increasing availability of free high-frequency financial data online, the opportunities to implement these techniques to forecast the behavior and manage assets arise. Taking advantage of this available information for decision-making is key to build a great portfolio and decide asset allocations.

In this project I use machine learning algorithms, in particular recurrent neural networks, to forecast the returns of assets of a defined asset universe that conform a portfolio. With this forecast, I then optimize the portfolio weights and decide the allocation on each asset, creating the future optimal portfolio.

The objective…


1. Introduction

A very important lesson we learn early on in our lives is the importance of saving and investing for our future. This should be a way to guarantee a good quality of life for our later years and even a good source of income and financial security.

Choosing in what to invest is a very important part of the process, because there is a wide variety of options for very different risk profiles. And as small investors, we face additional restrains.

These restrains could be related to the size of the capital we manage at the moment (which limits the…

Alejandro Duenas Garcia

Economist with focus in finance and data science

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store