The purpose of the master’s thesis is to compare and analyze momentum strategies using a broad selection of futures contracts from equity, bond and commodity asset classes. Momentum can be seen as a pervasive asset pricing anomaly and it is found in academic studies to exist in practically all major asset classes. Momentum challenges directly efficient-market hypothesis and currently there is no agreement as to the underlying cause of momentum.
Literature in momentum divides into two approaches: cross-sectional- and time-series momentum. The former was introduced by Jegadeesh and Titman (1993) and the latter by Moskowitz, Ooi and Pedersen (2012). Momentum can be applied in several different forms and whereas most studies in momentum applies only one of the two approaches, this study examines both by examining the most conventional strategies from the cross-sectional- and time-series momentum literature. Also volatility weighed and linearized versions of momentum are examined.
We analyze the performance of momentum strategies by summary statistics and also through risk adjusted performance. Strategy returns are regressed against conventional factor models, which include momentum factors. Volatility weighted versions of momentum strategies are more capable of producing better risk adjusted returns than regular or linearized versions. From strategy specifications, generally the 12-month lookback and 1-month holding strategy brings higher returns that 6-month lookback and 6-month holding alternative. These results are also presented on asset class level and it can be seen that momentum strategies are more successful when aggregated on all assets.
The study also examines market conditional performance during different economic cycles. No statistically significant difference can be seen for momentum returns during different economic cycles. Also momentum performance in comparison to market return is analyzed and we a see peculiar momentum “smirk” for regular strategies and more typical momentum “smile” for volatility weighted strategies. Momentum strategies returns cannot be seen as compensation for crash risk.
A decomposition of momentum returns shows that the sources of the returns are different for each of the approaches. However, cross-sectional and time-series momentum share also similarities — they have similar exposures against conventional momentum factors as well to one another and cross-sectional asset-class specific momentum strategy returns can explain the returns of corresponding asset-class specific returns from time-series momentum.
This study provides information on how asset managers use momentum and what are the practical implications of applying these strategies on futures contracts. The strategies presented apply to other asset classes as well. Compared to traditional 60/40 portfolio, addition of momentum can improve expected returns and risk/return tradeoff. Also as individual investing strategy, momentum is viable option to enhance the performance assets and portfolios.
| Identifer | oai:union.ndltd.org:oulo.fi/oai:oulu.fi:nbnfioulu-201611102990 |
| Date | 14 November 2016 |
| Creators | Depner, P. (Petri) |
| Publisher | University of Oulu |
| Source Sets | University of Oulu |
| Language | English |
| Detected Language | English |
| Type | info:eu-repo/semantics/masterThesis, info:eu-repo/semantics/publishedVersion |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess, © Petri Depner, 2016 |
Page generated in 0.0148 seconds