Alternative Neural Nets Approaches for Enhancing Stock Picking using Earnings Forecasts

In the last decade, neural networks have drawn noticeable attention from many computer and operations researchers. Essentially, a Neural Network is a non-parametric estimation technique. It does not make any distributional assumption regarding the underlying variable. In various studies Artificial Neural Network (ANN) models, have been proved to be powerful predictive tools, where a variable is explained by a set of explanatory variables without assuming any structural or linear relationship among the variables. Some critical success factors to train ANN are the network architecture, network design algorithm, training algorithm, and stop training conditions. While some previous studies, have found encouraging results with using this artificial intelligence technique to predict the movements of established financial markets, there is a lack of studies examining the stock picking ability of different ANN paradigms, taking into account analyst earning forecasts as input data. Our approach is based on the notion, that trading strategies guided by forecasts of stock earnings, may be effective and lead to higher profits. This paper attempts to enhance the stock selection process by employing ANN to select stocks in the Us stock market. Neural networks are used to identify stocks for the portfolio which are likely to outperform the market, given the forecast earnings information of stocks. Our purpose is to compare various ANN models, to identify critical predictors to forecast stock prices and to see, which model show the best stock picking ability, increasing in this way investment strategy profitability for the professionals in the market. The competiting models, are examined in terms of various trading performance and economic criteria, like annualized return, Sharpe ratio, maximum drawdown, annualized volatility, average gain/loss ratio etc via a trading experiment. Empirical experimentation suggests that by using artificial neural networks for nonlinear predictions there is potential economic value for subsequent portfolio choices. ANN-based investment strategies, once financial analyst earning forecast are considered, obtain higher returns than other investment strategies examined in this study. Consequently, we find that the returns obtained from the equally weighted portfolio formed by the stocks selected by neural networks, outperform those generated by the buy and hold strategy, computed with a benchmark index for a given period under investigation. The influences of the length of investment horizon and the commission rate are also considered. The present study does not support efficient market hypotheses.