| Abstract [eng] |
THE ROLE OF LEADING INDICATORS IN PREDICTING VOLATILITY IN FINANCIAL MARKETS TOMAS ZYKAS Master‘s Thesis Master’s study program, finance and banking Vilnius University, Faculty of Economics and Business Administration Supervisor – doc. dr. Greta Keliuotytė - Staniulėnienė Vilnius, 2025 SUMMARY 70 pages, 12 figures, 8 tables, 78 references, 10 appendices. The main goal of this master's thesis is to determine whether the use of leading indicators has a statistically significant impact on forecasting the European stock market. The thesis is divided into three main parts: theoretical aspects of using leading indicators to predict stock market returns/volatility, the methodology for applying leading indicators in stock market volatility forecasting, and the corresponding empirical study. In the first chapter, the types of economic indicators are defined, highlighting the benefits of each economic indicator. The most frequently recurring leading indicators in studies are also identified, along with their benefits. Subsequently, the most common challenges in the use of leading indicators are outlined, as well as frequently conducted economic studies that utilize leading indicators in forming stock market forecasts. Following the theoretical overview, the thesis describes the process of the forecasting study. While creating the forecasting model, 6 different leading variables were used, treated as independent variables, to determine their impact on the dependent variable: the volatility of returns of the EUROSTOXX 50 stock package. To assess whether leading indicators influence volatility forecasts, ARMA and GARCH models were constructed following the defined research methodology. The results of the forecast studies indicate that incorporating leading indicators into simpler models, such as ARMA, can provide statistically significant information for predicting the direction of volatility changes. However, when leading indicators are included in more complex GARCH models, this effect diminishes due to the excessive complexity of the model. Conclusions and recommendations were made based on the information mentioned in the theoretical section and the predictive studies conducted. |
