Alternative Neural Network Approaches for Enhancing Stock Picking Using Earnings Forecasts

Abstract

Interest in financial markets has increased in the last couple of decades, among fund managers, policy makers, investors, borrowers, corporate treasurers and specialized traders. Forecasting the future returns has always been a major concern for the players in stock markets and one of the most challenging applications studied by researchers and practitioners extensively. Predicting the financial market is a very complex task, because the financial time series are inherently noisy and non-stationary and more it is often argued that the financial market is very efficient. Fama (1970) defined efficient market hypothesis (EMH) where the idea is a market in which security prices at any time ‘fully reflect’ all available information both for firms’ production—investment decisions, and investors’ securities selection. Furthermore, in EMH context no investor is in a position to make unexploited profit opportunities by forecasting futures prices on the basis of past prices. On the other hand, a large number of researchers, investors, analysts, practitioners etc. use different techniques to forecast the stock index and prices. In the last decade, applications associated with artificial neural network (ANN) have drawn noticeable attention in both academic and corporate research.