Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Master's Thesis |
| Scope | Discipline-based scholarship |
| Title | Who is on the Other Side? |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Business, Economics and Informatics |
| Number of Pages | 88 |
| Date | 2020 |
| Zusammenfassung | Every market must have a buyer and a seller. Understanding the motivations behind every investor on modern equity markets, their performance and interactions is an impossible task as it involves knowing their investment strategies and personal wealth. However, if we move this problem from the universe of individual investors into the factor universe, said task becomes doable and the objective of this thesis. This thesis examines interaction of market anomalies, commonly known as factors, based on a model of dynamic interaction of factors: the evolutionary portfolio theory model of Evstigneev et al (2016). Its main goal is to extract relative factor wealth and use it to study the behavior of investors funding those investment styles. To do this, we first construct selected list of factors using a novel approach where we use US sector indices data as our main data source. This way, we drastically reduce the total amount of data and any issues with data corruption, unavailability and any biases we may introduce during the data cleanup and preparation stage. Furthermore, by directly constructing the factors in every time period we obtain the investment strategies of our agents in this framework. Signals used for factor construction mimic those used in the literature. We compare our factor returns to those calculated by Kenneth French, in Fama and French manner, to show the degree of similarity between the two. We replicate the findings of Asness et al (2013) and Novy-Marx (2013), regarding the factor returns relationship. Theoretical basis behind our investigation allows us to show how relative factor wealth of all factors, save Market, must average at zero. This comes from the long-short construction principle predominantly used in the literature. We interpret relative factor wealth as a proportion of total capital in market following the factor determined investment style. This allows relative wealth to become negative, which is necessary due to aforementioned constrain, which implies that a given fraction of capital is invested into a strategy contrarian to the one factor describes. Having extracted the relative wealth, we are able to deepen the analysis of Asness et al (2013) and Novy-Marx (2013) by examining, not only returns, but also factor investment strategies and relative wealth of the relevant factors. Individual analysis of factor wealth includes first examining the wealth dynamics during different market condition and regimes. Domestic events appear to be more impactful, as compared to the global ones. We find both the expected results such as flight to safety during turbulent times, and high attractiveness trend-following approaches and aggressive high volatility growth companies during the recovery period, as well as some less expected such as the recent resurgence of using high profitability as the key company characteristic, or the apparently persistent overall popularity of value companies. Secondly, we find predictability in individual series, implying a degree of inertia behind the invested capital. Majority of autoregressive terms are estimated within a narrow band implying a general, factor independent, speed of capital mobility in the market. Lastly, we find weak-to-nonexistent evidence of adaptive behavior behind factor investing. Adaptive behavior would imply that relative wealth is sensitive to the excess market return of the factor. |
| Export |
BibTeX
|