Statistical Arbitrage in the Egyptian Stock Market (An Applied Study)

Abstract

Statistical arbitrage is a short-term equity trading strategy that exploits parametric and nonparametric statistical techniques to identify the relative mispricing between various securities, thereby generating positive returns while hedging risks. The idea evolved from pairs trading, which is dependent on identifying a portfolio of two stocks that moves together in some historical manner. The strategy is to attain positive returns from the temporary deviations that may occur in their equilibrium prices by implementing a long-short (buy-sell) position for the two stocks in a pair. Trading decisions are automatically generated using trading algorithms. This in turn eliminates human subjectivity in taking decisions and enables the consideration of numerous portfolios and trades. With the advancement of statistical techniques and computing power, not only a pair of stocks may be observed, but hundreds or even more can be monitored and combined in an arbitrage portfolio. These pairs can be chosen by sectors, industries, or advanced clustering techniques to effectively hedge numerous risk factors while generating positive returns. The general objective of this study is to uncover potential arbitrage opportunities present in the Egyptian stock market (EGX) by constructing trading algorithms based on combining different statistical techniques and investigating whether the risk and return of these techniques can be improved by implementing a dynamic trading threshold.