Stock Selection via Nonlinear Multi-Factor Models

Part of Advances in Neural Information Processing Systems 8 (NIPS 1995)

Bibtex Metadata Paper

Authors

Asriel Levin

Abstract

This paper discusses the use of multilayer feed forward neural net(cid:173) works for predicting a stock's excess return based on its exposure to various technical and fundamental factors. To demonstrate the effectiveness of the approach a hedged portfolio which consists of equally capitalized long and short positions is constructed and its historical returns are benchmarked against T-bill returns and the S&P500 index.