Integrating Artificial Neural Network Models By Markov Chain Process: Forecasting The Movement Direction Of Turkish Lira Us Dollar Exchange Rate Returns

In this study we first modeled daily US dollar returns as the discrete state Markov chain process and second we trained an Artificial Neural Network ANN model in order to estimate direction of dollar return The trained model provides valuable information about the direction of next day return Keywords: Artificial Neural Networks Markov chains Conditional probability Exchange rate returns Jel: C45 C53 F31