Forecasting the Direction of BIST 100 Returns with Artificial Neural Network Models

Abstract

In this paper, Artificial Neural Networks (ANN) models are used to forecast the direction of Borsa Istanbul 100 (BIST100) index returns. Weekly time-lagged values of exchange returns, gold price returns and interest rate returns are used as input to ANN models in the training process. Results of the study showed that BIST100 index returns follow a specific pattern in time. Estimated ANN models provide valuable information to the investors and that BIST100 stock market is not fully informational efficient.