This paper investigates the model estimation and data forecasting of exchange rate using\nartificial neural network. Recent studies have shown the classification and prediction power of\nthe neural networks. It has been demonstrated that a neural network can approximate any\ncontinuous function. Here, in a technical approach, it has been used ARIMA and neural\nnetwork for a short-term forecast of daily USD to Rial exchange rate. ANN is employed in\ntraining and learning processes and thereafter the forecast performance measured making use\nof two common loss functions. The comparison demonstrates that neural network is far better\nthan ARIMA, the error is about the half.\nThereafter, in a fundamental approach via another neural network the effects of some of the\nmost important economic variables on exchange rate prediction in a long-term sense are\nstudied. By sensitivity analysis, the importance and the weight of each economic variable on\nexchange rate has determined. The results show that it is possible to estimate a model to\nforecast the value of exchange rate even by having access to a limited subset of data.
Loading....