Testing the Performance of Simple Moving Average With the Extension of Short Selling

Abstract

In this thesis, we test the performance of market timing based on simple moving average, which is one of the most popular trading strategies used by investors and practitioners to date. Previous studies have found evidence both in favour and against the effectiveness of the strategy, while a definite conclusion is yet to be commonly recognized. To address this, we reassess a previous study done on US portfolios with stocks from the NYSE, AMEX and NASDAQ, further investigate the effectiveness of the strategy in Norwegian portfolios constructed by stocks from the Oslo Stock Exchange (OSE). This thesis contributes with a new extension, possibly for the first time, testing the moving average strategy with short selling the underlying portfolio when triggered a sell signal. We use value-weighted portfolios with monthly returns from both the US and Norwegian market sorted by size, book to market and momentum. Our results revealed both lower risk and return in general by the moving average strategy compared with buying and holding, providing no evidence supporting superior performance of the strategy in neither US nor Norwegian portfolios. Shorting the underlying portfolio showed similar results, however, one interesting finding is the behaviour of the short strategy, which tend to amplify the normal simple moving average strategy’s performance.