Modeling the Impact of Mass Media Communication on Stock Trading Decisions

<p>This paper is dedicated to analysis of the mass media&rsquo;s impact on the stock market. The main goal of the research is to identify the significant connections between news flows and dynamics of the equity market.</p><p>Empirical base consists of historical monthly data of financial standing of five largest U.S. companies in high technology: Apple, Microsoft, IBM, Oracle, Intel. The initial sample contains average closing price of the share over the month, the amount of declared dividends and the total trading volume of the New York Stock Exchange for each above mentioned corporation.</p><p>The basic concepts of the theory of financial bubbles are examined in the first part of this study, including: definition, classification of economic bubbles and a brief review of the literature on the subject. Further we examine the impact of information flows on the stock trading decisions based on real events.</p><p>The second chapter begins with the description of simple capital asset pricing model. Then we scrutinize two regime-switching models of speculative behavior. The underlying model of this paper is three-regime model of the behavior of stochastic bubble described in the work of Brooks and Katsaris (2005). This chapter concludes with the construction of new variable, that represents the role of information events on the securities market. The introduced explanatory variable indicates the frequency of requests about financial condition of the company in the web search engine Google, which is one of possible &nbsp;indicators of the potential investors&rsquo; interest.</p><p>In the third chapter we present the results of an evaluation based on the maximum likelihood method and the interpretation of the results for the original model with three states for five companies and the modified model with respect to the influence of the news flows on the stock market only for Apple Inc.</p><p>Finally we describe the main results and recommendations that could be useful for further application of the materials and conclusions of this work.</p><p>In the process of the this study the modified model has proven to be sufficient and more accurate to the initial data in comparison with the original three-regime model. This allows us to draw a conclusion about the significance of the impact of information events for analysis of the stock market behavior.</p>