| Abstract [eng] |
Analysis of relationship between stock market and macroeconomic indicators remains relevant and is widely discussed among scientists in relation with constantly changing economic situation, unprecedented events and increasing availability of historical information. Situation of the United States of America is usually analyzed since changes in the US market has a significant impact on stock market and economic indicators of other countries. In most cases, indicators are analyzed on a countrywide basis, however, market analysis on a sectoral level might imply inequal impact on stock market as well as provide additional insights for potential investors. The aim of this project – identify and compare relationship between macroeconomic indicators and stock market at the sectoral level in the United States of America. Analysis of previous scientific research in the first part of the project revealed that findings of research on the relationship between stock market and macroeconomic indicators are manifold. Nonetheless, most findings confirm the existence of relationship between the aforementioned variables considering stock market indicators as leading compared with the economic ones. Economic situation in the country is usually described by composite leading indicator, interest rates, unemployment rate, index of industrial production and consumer price index while stock market is described by a stock index. Process of the research as well as methods used are presented in the second part of the project. In the third part of the project, empirical analysis is carried out for the period of 1996–2021. GARCH type models have been applied to the sectoral US stock market data after stylized facts were confirmed for the time series. Results of the univariate analysis of sectoral stock market confirmed the assumption of historical values’ impact on current values. In addition, the significance of shock asymmetry to the time series as well as differences between sectors have been proved. Meanwhile analysis of results of multivariate vectoral autoregression led to the conclusion that there is a relation between sectoral stock returns and macroeconomic variables. Consumer price index differ from the remaining economic indicators as no Granger causality has been detected with consumer discretionary, communication services, consumer staples and financials sectors. Information technology sector distinguishes due to one-way causality on macroeconomic indicators. Taking results of structural analysis into account, the effect of impulses on other variables is considered to be long-term with consumer staples sector shock being least significant. Meanwhile the most significant impact on the forecast error variance has the time series themselves, the influence of sectors on the economic variables does not exceed 20 %. |
