Abstract:

Today, banks that come to the top of the financial sector are working on renewing themselves and openly adapting to innovations over the day. The banking sector is also in an important position for our country’s economy. One of the main objectives of the banks in their activities is to grow byining sustainable profitability and efficiency. In this study, the active sizes of the banks were taken into account. However, the aim is to make profitability forecasts by the method of artificial nerve networks of the large-scale banking groups. The training of the artificial nerve network model in the bank’s profitability forecast has been solved with the Matlab (Version 7.12) program. The application has taken into account the three-month period data between March 2003 and June 2017. However, the average self-source profitability is used as a dependent variable, and other data are treated as an independent variable, calculating the average absolute percentage error (MAPE) ratio between test data and real data. The estimated values of the artificial nervous network model and the estimated values of the test data have been observed close to each other.  Therefore, it has been made forecast with the networks. As a result, artificial nerve networks have been shown to be a method that can be used in future prediction studies. At the same time, the variables used and the findings obtained have shown that the bank profitability can be predicted forward.

Keywords: