Investigation Into The Effectiveness Of Long Short Term Memory Networks For Stock Price Prediction

TL;DR

This paper evaluates the effectiveness of various LSTM network architectures trained with backpropagation through time for stock price prediction, providing insights into their predictive capabilities.

Contribution

It systematically constructs, trains, and tests different LSTM architectures to assess their performance in stock price prediction tasks.

Findings

LSTM networks can effectively model stock price data.

Different architectures show varying prediction accuracies.

The study highlights the potential of LSTM for financial forecasting.

Abstract

The effectiveness of long short term memory networks trained by backpropagation through time for stock price prediction is explored in this paper. A range of different architecture LSTM networks are constructed trained and tested.