Performance and Risks Using Current Business Results and News Sentiment Analysis

In a global economy, the banking industry is subject to a variety of factors, both internal and external. Internal factors include current operating results, internal management decisions, customer policy, etc. External factors include changes in macroeconomic indicators, political events, and the competitive environment. The description of factors that have a direct impact on the results of banks can be found in the news from the media. Therefore, the problem of quantitative measurement of the impact that news has on the performance of banks becomes acute. The objective of this paper is to develop a model that combines bank performance data and the news background, which can predict the profitability and risk performance of bank institutions that are important for investment decisions. During the study Russian banks’ performance data was collected, as well as economic and political news from Lenta.ru. The target variables for model fitting are the following indicators: default event, share price, ROE, ROA, and capital adequacy. As a result of the research, a collection of models was created with effective forecasting of banks' profitability and risk indicators. At the same time, different models and different sets of data for training had the highest accuracy for each of the indicators. Thus, it was concluded that for effective forecasting of future results of the bank, it is necessary to evaluate the obtained forecasts in the context of each indicator separately, and therefore to create a different model for each of them. The results obtained can be used as new features for combined models that predict the overall value of the bank, as well as to make correct management and investment decisions about different aspects of the bank's activities based on individual indicator predictions.