This is a quant finance experiment at replicating the results from "Enhancing Time Series Momentum Strategies using Deep Neural Networks" by Bryan Lim, Stefan Zohren, and Stephen Roberts. Their works uses 4 different DNN architechtures to backtest returns from 88 future contracts over a period of 20 years from 1995 - 2015. The paper can be viewed here: https://arxiv.org/pdf/1904.04912
The primamry DNN architectures used are Single Lasso Regression, Multilayer Perceptron, Long Short Term Memory, and WaveNet. This project focuses just on recreating results by using the MLP architecture. Hopefully a sufficient backtest of the results will simulate a similar result to that of the paper.