Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Master's Thesis |
| Scope | Discipline-based scholarship |
| Title | Cross-Sectional Variation in Commodity Futures Risk Premia: An Analysis based on Factor Models |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Business, Economics and Informatics |
| Number of Pages | 50 |
| Date | 2018 |
| Zusammenfassung | A sharp increase in the popularity of commodity investing has triggered an unprecedented inflow of institutional funds into commodity markets in the past decades. This development, which is known as the financialization of commodity markets, led to the emergence of commodities as an individual asset class. Consequently, the understanding of the risk premia structure of such investments is essential, as it strongly impacts the decision making process of the corresponding market participants. This Master Thesis aims to assess the most systematic risk factors of commodity futures contracts within a cross-sectional asset pricing setting and under consideration of structural changes over time. The fundamental dataset of this Master Thesis consists of 32 commodity futures contracts, which could be group into five sectors (Energy, Grains, Livestock, Metals and Softs), starting in January 1984 and lasts till August 2017. Additionally, the author built a broad set of risk factors which are related to the term structure of the commodity futures curve, trading strategies, hedging pressure and macroeconomic fundamentals. Based on this data a four-factor asset pricing model has been developed that is capable of reconciling more than three-fourths of the entire risk premia of an aggregated commodity futures portfolio. The model has been estimated by applying the well-known Fama-MacBeth procedure. Moreover, the robustness of this four-factor model was cross-checked and confirmed with the generalized method of moments (GMM) approach. The achieved estimates for the commodity futures monthly risk premia are -0.11% for the term structure factor (basis), 0.66% for the active trading strategy (carry), 0.28% for hedging pressure and finally, 0.17% for the macroeconomic risk factor (OECD industrial production). What does these results imply for practitioners? Well, as long as they do not follow an active trading strategy, their positions are mainly exposed towards the hedging pressure and macroeconomic risk factors. These factors have to be monitored very closely by looking at the corresponding indicators in order to notify portfolio relevant events as soon as possible. Additionally, they have to be very clear about their beta sensitivities towards these risk factors, as those are directly linked. However, if they are engaged in an active trading strategy, such as momentum or carry, a much more powerful risk premia comes into play which leads to higher expected returns, on average, as a compensation for the additional risk. These considerations are even more important when it comes to commodities from the energy or metal sector, as they report a much stronger dependency to this risk factors. Having all this in mind an active commodity portfolio management can be mastered without taking too much unseen risk premia into account. An emerging literature dates the start of the financialization of commodity markets around the year 2004, showing changes in the behaviour of futures prices and their volatilities. Hence, the second part of this Master Thesis aims to identify quantitatively such structural changes on both, the commodity sector level and the aggregated portfolio view. By applying an extended version of the cumulative sum (CUSUM) approach one can identify a significant break in the statistical distribution of the mean for the energy and metal sector around the year 2002. This time point was used to split the entire dataset into two subsamples which have been evaluated again with the four-factor model using the Fama-MacBeth procedure. Thereby, one can conclude that the risk premia for the term structure and macroeconomic factors are relatively identical within these two subsamples and hence not time-varying. However, this is not true for the other two risk factors. The risk premia for the carry factor is remarkably higher in both subsamples than in the entire dataset. This indicates that active trading strategies in the commodity markets are much more time-varying in short time periods and thus bears an increased amount of risk potential. On the other hand, the risk premia for hedging pressure turned even into the negative area. This could intend that the original hedging pressure theory no longer holds, as institutional investors enlarge their financial engagement in the commodity markets. Finally, one can note that the estimated risk premia for the second subsample show much higher standard errors which could be seen as a clear sign for an increased volatility in the commodity futures markets since the year 2002, at least in the energy and metal sector. |
| Export |
BibTeX
|