Stock Price Forecasting and Hypothesis Testing Using Neural Networks

TL;DR

This paper explores the use of neural networks for stock price prediction across major exchanges and employs the results to test the efficient-market hypothesis through statistical analysis.

Contribution

It introduces neural network models for stock forecasting and applies them to evaluate market efficiency with formal statistical testing.

Findings

Neural networks can predict stock prices with certain accuracy.

Data normalization impacts prediction performance.

Results challenge the efficient-market hypothesis.

Abstract

In this work we use Recurrent Neural Networks and Multilayer Perceptrons to predict NYSE, NASDAQ and AMEX stock prices from historical data. We experiment with different architectures and compare data normalization techniques. Then, we leverage those findings to question the efficient-market hypothesis through a formal statistical test.