Empirical analysis of machine learning techniques for prediction of indian exchange rate

Abstract

Throughout the past few decades, there has been a dramatic surge in the currency market. The changes play an important role in balancing the market’s characteristics. As a result, accurate change price forecasting is essential to improve the success rate of many businesses and fund managers. Despite the fact that the market is renowned for its erratic behavior and volatility, there are organizations like agencies, banks, and others. In order to estimate the extraneous interchange rate of the dollar against the rupee with a high degree of accuracy, we used three distinct types of methodologies in this article. This research uses three different types of neural network models: ANNs (Artificial Neural Networks), LSTMs (Long Short-Term Memory Networks), and GRUs (Gated Recurring Units). The results depict that GRU’s model is outperforming the other two models.