VALUTAZIONE DIREZIONALE E CROSS SECTION DI STRATEGIE BASATE SUL MOMENTUM

In literature, there exist a lot of works aimed to investigate the performances and profitability of different investment strategies. Carry, value and momentum are among those strategies that give back better results during time, but does not exist the same mole of literature that estimates and compares their application, among the same sets of different asset classes. This dissertation arises from the idea of replicating a technical paper that investigates the performance of three popular strategies such as carry, momentum, and value in different implementation: cross-sectional and time-series. The title of the cited paper is: "Dissecting Investment Strategies in the Cross Section and Time Series", by Jamil Baz, Nick Granger, Campbell R. Harvey, Nicolas Le Roux, and Sandy Rattray. Their goal was to find which are the market conditions that favor mostly the application of a particular setting. Although the literature about different investment strategies was huge - Asness, Moskowitz, and Pedersen (2013) and Moskowitz, Ooi, and Pedersen (2012) - they noticed what we have already expressed so far. Those papers were limiting the analysis only to some strategies or some asset classes. Therefore, they decided to fill this gap in 2015 by analyzing and comparing the relative performance of both cross-section and time-series approaches, as well as combined, among a vast number of asset classes: equity, currencies, commodities, and fixed income. They showed how those strategies were profitable among their sample, resulting in the best performances when combined. However, the real focus is not to demonstrate which strategy is more profitable but rather to understand the market conditions where a particular setting has the best chance to work. The following dissertation focuses on momentum strategy, aiming to come up with very close and similar results, among the same asset classes considered in the survey, building a specular sample and following the same steps. The project will be considered successful if it will end up with consistent results between 1990 and 2015, letting the possibility of extending the analysis to further years.