Can Deep Machine Learning Outsmart the Market? a Comparison Between Econometric Modelling and Long- Short Term Memory
Year: 2016
Type: article
Abstract: Using long-short term memory (LSTM) recurrent neural network (RNN) architecture, we analyse data from the Romanian stock markets in the attempt to forecast its future trend. Then we try to compare the results using the classical statistical modelling tools, further employing back testing to prove our findings. We believe that the LSTM should be the next tool in balancing portfolios and reducing market risk.
Authors: Eva Dezsi, Ioan Alin Nistor
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2747096140
