Stock returns forecast: an examination by means of Artificial Neural Networks

TL;DR

This paper investigates the ability of artificial neural networks to predict intraday stock returns in the Brazilian market, revealing that stock size influences predictability, with larger stocks being less predictable.

Contribution

It applies neural networks to analyze stock return predictability in an emerging market, highlighting the relationship between market capitalization and predictability.

Findings

Larger stocks are less predictable than smaller stocks.

Neural networks can model intraday stock return predictability.

Predictability varies with market capitalization.

Abstract

The validity of the Efficient Market Hypothesis has been under severe scrutiny since several decades. However, the evidence against it is not conclusive. Artificial Neural Networks provide a model-free means to analize the prediction power of past returns on current returns. This chapter analizes the predictability in the intraday Brazilian stock market using a backpropagation Artificial Neural Network. We selected 20 stocks from Bovespa index, according to different market capitalization, as a proxy for stock size. We find that predictability is related to capitalization. In particular, larger stocks are less predictable than smaller ones.