The dynamics of the financial market depends on the expectations of investors, which are largely determined by economic and political events. To form investment strategies, it is important to understand which events described in the news may affect changes in the value of assets. The purpose of this work is to identify the topics of political news that affect the profitability of shares of Russian companies, and to compare the predictive power of news from Russian and foreign sources. This study investigates the relationship between political news about Russia, obtained from domestic (“Interfax”) and foreign (“New York Times”) sources, and the stock returns of 193 Russian companies over the period from September 1, 2021 to August 31, 2023. There were 30 political dictionaries from each source identified using the Latent Dirichlet Allocation model, and the differences in the highlighted political themes were noted. Time-series models were used to test hypotheses about the impact of political news on the stock prices of Russian companies. The study demonstrates that the Russian stock market’s dynamics are impacted by news from various sources. Specifically, political dictionaries derived from foreign sources enhance the return predictions for the stocks of 142 Russian companies, whereas those from domestic sources improve the forecasts for 146 ones. Nevertheless, models incorporating political news from domestic sources yield higher-quality return forecasts. Additionally, using the Random Forest algorithm, it is demonstrated that the domestic media’s interpretation of events, which are covered in both Russian and foreign news, exerts a more substantial influence on the domestic stock market. Furthermore, models that integrate political dictionaries from both sources exhibit superior quality compared to those that rely on news from a single source. Based on the results obtained, it is demonstrated that incorporating political news into investment decisions enables investors to construct stock portfolios with higher returns
Keywords
political news, Russian stock market, foreign and domestic media, Latent Dirichlet Allocation, comparative analysis