Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Master's Thesis |
| Scope | Discipline-based scholarship |
| Title | Hierarchical Clustering Method for Country Equity Allocation |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Business, Economics and Informatics |
| Number of Pages | 40 |
| Date | 2019 |
| Abstract Text | This thesis applies graph theory and a new machine learning method, Hierarchical Clustering, to solve strategic asset allocation problem. The major investment context is equity portfolio allocation on the country level. This thesis stands from the view point of a global investor and all countries in MSCI World Index have been taken into account. The portfolio constructed by Hierarchical Risk Parity (HRP) method is compared to traditional Minimum Variance Portfolio (MVP) and Inverse Variance Portfolio (IVP) in out-of-sample backtests. Market portfolio is measured as the benchmark portfolio. Different lookback periods (5 years/3 years) and rebalancing periods (1 year/6 months) have been taken into consideration. T-tests are conducted to check whether the average returns are significantly different. The study also goes further to investigate the how HRP performs in developed and emerging markets separately. This thesis provides the first attempt of applying HRP algorithm to country equity allocation. The out-of-sample backtests results show that HRP delivers better risk-returns performance compared to traditional MVP and IVP methods. However, the HRP performance still cannot beat the market portfolio. The bull market trend of S&P 500 in the last 10 years and underweight of the U.S. equity market of HRP could possibly explain this observation. |
| Export |
BibTeX
|