Stock Market and Macroeconomic Variables: VAR and PVAR Approach

Matěj Doležal

Master's thesis

Abstract:

The study aimed to examine the relationship between the stock market and the real economy. The analysis was conducted based on three models: the S&P 500 stock index, the FTSE 100 stock index, and subsequently, a panel data structure. The observed period chosen was 1Q05-4Q20. The Vector Autoregression (VAR) model was selected for analysis, with subsequent extension to a panel version (PVAR). The macroeconomic variables included inflation, GDP, interest rates, FX rates, and even some market variables such as oil prices and the VIX index. The study concludes that the stock market index generally precedes developments in the real economy across all three models, while the reverse causality was not verified. Furthermore, it was found that there is no significant difference between individual countries, and the results are very similar across all models. Generally, including all variables appears appropriate, as all exhibit statistically significant connections with the stock market. Additionally, it is noted that the panel dataset did not yield any new insights; however, it is important to note that the scope of this dataset is relatively small and focuses only on developed countries, which is subject to criticism.

Abstract:

The study aimed to examine the relationship between the stock market and the real economy. The analysis was conducted based on three models: the S&P 500 stock index, the FTSE 100 stock index, and subsequently, a panel data structure. The observed period chosen was 1Q05-4Q20. The Vector Autoregression (VAR) model was selected for analysis, with subsequent extension to a panel version (PVAR). The macroeconomic variables included inflation, GDP, interest rates, FX rates, and even some market variables such as oil prices and the VIX index. The study concludes that the stock market index generally precedes developments in the real economy across all three models, while the reverse causality was not verified. Furthermore, it was found that there is no significant difference between individual countries, and the results are very similar across all models. Generally, including all variables appears appropriate, as all exhibit statistically significant connections with the stock market. Additionally, it is noted that the panel dataset did not yield any new insights; however, it is important to note that the scope of this dataset is relatively small and focuses only on developed countries, which is subject to criticism.

Keywords

Stock Market, VAR Model, Macroeconomic Variables

Keywords

VAR Model, Stock Market, Macroeconomic Variables

Language used: English

Date on which the thesis was submitted / produced: 24. 5. 2024

Thesis defence

• Date of defence: 18. 6. 2024
• Supervisor: Karel Brůna
• Reader: Žaneta Tesařová

Vysoká škola ekonomická v Praze

Master programme:
Finance