Hodnocení vlivu sentimentu na kurzy akcií

Abstract

The thesis is focused on the evaluation of the influence of sentiment on stock prices. The aim of the thesis is to verify whether the frequency of searching for selected keywords with a negative connotation in connection with names of selected companies using Google web search engine may have an impact on stock prices of the companies. The thesis is divided into five chapters, where the first chapter is an introduction and the fifth chapter is a conclusion. The second chapter is devoted to the theoretical background and historical development of behavioral finance. The second part of the second chapter is devoted to the behavioral heuristics and biases. The third chapter describes econometric and statistical methods needed to verify accuracy of regression models frequently used in the fourth chapter. The second part of the third chapter describes methods of measuring sentiment and their basic division, which can be used as an alternative to financial or fundamental analysis. The fourth chapter describes process of acquisition and editing data used in the regression models, process of creating regression models and their subsequent modifications. Overall, it can be stated that sentiment captured through the frequency of keyword searches can have a statistically significant influence on the stock prices, although rather low.