Exchange Rates’ Change by Using Economic Data with Artificial Intelligence and Forecasting the Crisis

• Published in: Procedia computer science 2019, Vol.158, p.316-326
• Main Authors: Güler, Kemal, Tepecik, Abdulkadir
• Format: Article
• Language: English
• Subjects: Artificial Neural Networks; Crisis Forecast; Currency; Forecasting; Gold Rate
• ISSN: 1877-0509; 1877-0509
• DOI: 10.1016/j.procs.2019.09.057

Artificial neural networks have recently been widely used in the field of finance as well as in every field. Exchange rates and gold prices are vital for banks, stock exchanges and businesses. Therefore, it is important for such organizations to make accurate estimates of foreign exchange and gold exchange rates. There are many studies that can affect and predict gold prices and exchange rates in the literature review. In this study, it is aimed that the fluctuation in foreign exchange and gold exchange rate can be estimated by using artificial intelligence methods and the forecast results can be used in predicting crisis. In the study, monthly data between 2006 and 2018 were used for USDTRY exchange rate estimation. External factors that may affect the USDTRY exchange rate are added as an independent variable. BIST100 index data, US inflation data, inflation data for Turkey, Turkey American interest data and interest data were used as external factors. Monthly data between 2000 and 2018 were used to estimate gold prices. External factors that may affect gold prices are added as an independent variable. BIST100 index data, silver data, USDTRY exchange rate data and US inflation data were used as external factors. Annual data between the years 2000 and 2018 are used for the forecasting of the crisis. External factors that can trigger the crisis are added as arguments. USDTRY data rate, inflation data for Turkey, and Turkey import data rate data are used as external factors.