Discovering efficient techniques to enhance M&A prediction modelling methods

Abstract

Over the last two decades global economy and financial markets have seen many crises with an increasing frequency. Many of them were either unexpected or their effects were unpredictable. Therefore, a lot of conventional and popular investment opportunities have shortened or became less attractive for either private or institutional investors leaving a room for new instruments and products to gain popularity. Unlike new volatile and uncertain markets like cryptocurrency, there are potentially more stable processes such as M&A activity, which can be predicted to earn abnormal returns. In this paper, we study various financial and non-financial indicators of acquired and non-acquired companies to provide a set of variables that can describe a company from different perspectives. Next, M&A prediction model is designed. Then, techniques are discovered to increase its explanatory and predictive power, and flexibility making it applicable for different economic environments without being harder to implement it by a potential user. In the end, its efficiency is measured on a real data to compare it with a result of methods found in earlier papers.