Statistical Arbitrage & Fund Performance: An Empirical Analysis of Fund Returns

Abstract: Fund companies and banks argue that letting them manage one’s money is a wise decision. They argue that they are able to create substantial growth in value for the investor without requiring any other input than a small fee and an amount to be invested. This essay tests this claim in a two folded analysis. Time series data with historical value developments of 24 out of the 30 largest Swedish equity funds, along with the value development of the MSCI Sweden SEK stock market index, is used in the analysis. The first part of the analysis uses descriptive statistics of fund net returns and the benchmark index returns to assess whether or not fund managers are able to exploit statistical arbitrage opportunities in the Swedish equity market. It is concluded that fund managers are able to do this and consequently create average excess net returns that are greater than zero at an 80% confidence level. The second part of the analysis investigates if it is really worth the cost of fees to let an actively managed fund take care of one’s investment. The fact that fund managers are able to create positive excess returns is not good enough a reason to motivate fund investment on its own. It is investigated if the best three funds in the data set are able to outperform an active, but free, investment strategy that does not require financial sophistication. The investment strategy is based on a GARCH(2,1) model which is used to forecast the returns of the benchmark index and guide investment decisions. The second part of the analysis concludes that the funds are outperforming the proposed investment strategy on average, indicating that there is merit to the fund companies’ claim; it is worth the cost of fees to have one’s money looked after by an actively managed fund, given that the choice of fund is a well-informed decision.