Statistical Arbitrage on the Stock Markets

Statistical arbitrage is a relatively widespread investment approach that belongs among the so-called market neutral strategies. Although the first records date back to the 1980s, there is still no unified opinion on its ability to generate profits within the investors' community. The goal of the thesis is to introduce the theory of statistical arbitrage and consequently to test its functionality using real historical stock data. The thesis is divided into three main chapters. The first chapter establishes a theoreticalbase necessary to understand the following chapters. It begins with a brief introduction to the historical evolution of statistical arbitrage and states the basic principles of a market neutral strategies. It continues with the analysis of stock market movements using a statistical point of view and finally a theoretical background required for the application of this approach is laid down. The second chapter is fully devoted to statistical arbitrage and walks through the entire process of constructing a trading system. Primarily it deals with the statistical methods that form the core of the strategy and secondarily defines a few additional general aspects that are applicable to any investment approach. The final chapter focuses on testing the theoretical principles using real historical data. The test results should give us an idea whether it can be used to "beat" the market or not.