Utilising AI Models to Analyse the Relationship between Battlefield Developments in the Russian-Ukrainian War and Fluctuations in Stock Market Values

This study examines the impact of battlefield developments in the ongoing Russian–Ukrainian war, which to date has lasted over 1000 days, on the stock prices of defence corporations such as BAE Systems, Booz Allen Hamilton, Huntington Ingalls, and Rheinmetall AG. Stock prices were analysed alongside sentiment data extracted from news articles, and processed using machine learning models leveraging natural language processing (NLP). Although the main hypothesis was not confirmed due to methodological and data limitations, the study demonstrated that neural network-based models, specifically long short-term memory (LSTM) networks, effectively captured hidden temporal patterns. The model's performance was evaluated using root mean squared error (RMSE). Alternative models, including XGBoost, ARIMA, and VAR, were also tested but did not yield accurate forecasts. The findings highlight nonlinear patterns in the data and emphasise the importance of hyperparameter optimisation, such as tuning the number of epochs and LSTM layer sizes. Techniques such as Grid Search and Random Search significantly enhanced forecasting performance, resulting in stock price predictions with low RMSE. The full-scale war between Russia and Ukraine, which has been ongoing for more than 1000 days on European soil, triggered market reactions with the first shots fired on the Russian–Ukrainian border on 24 February 2022. Since then, Europe and other countries around the world have supported Ukraine militarily, humanitarianly, and economically while revising their defence doctrines and strategies, as well as increasing and modernising their military capabilities. News of Russian aggression and new orders for military products sparked market reactions, particularly changes in the stock prices of defence companies that ramped up production and continue to expand it. This study selected several of the largest defence corporations, whose stock prices showed significant responses to the full-scale Russian invasion of Ukraine, to examine whether other battlefield events elicit similar stock price reactions as the first reports. To investigate this, artificial intelligence tools, including several machine learning methods, were employed. The purpose of this article is to examine the impact of battlefield developments in the Russian–Ukrainian war on stock volatility among major players in the global defence industry, while also identifying the most effective method for forecasting stock prices by testing various machine learning (ML) models. NLP was used to capture sentiment and analyse news about the Russian–Ukrainian war. Vector AutoRegression, the XGBRegressor, the Long Short-Term Memory (LSTM) neural network, and the ARIMA model were used for stock price forecasting. The main hypothesis that stock price changes are influenced by battlefield developments in the Russian–Ukrainian war was not confirmed due to methodological and data limitations. However, the research results indicate that, in this case, the Long Short-Term Memory (LSTM) model is well-suited for time series forecasting, effectively capturing nonlinear time patterns and long-term dependencies for major defence corporations.