|
Cluster-based asset allocation strategies during market stress periods
Zacharová, Beáta ; Krištoufek, Ladislav (advisor) ; Čech, František (referee)
This thesis empirically examines the alternatives to traditional asset allocation strategies based on clustering mechanisms. Portfolio selection strategies utilizing hierarchical clustering are compared to the market benchmark and traditional methods: minimum-variance and equally weighted allocation, focusing on market stress periods. The allocation strategies are tested on daily stock prices of the S&P 100 index constituents from 2005 to 2021. The performance of Hierarchical Risk Parity (HRP) and Hierarchical Equal Risk Contribution (HERC) portfolios is evaluated across several market stress periods, including the financial crisis of 2007-2008 and the global coronavirus (COVID-19) pandemic in 2020. Empirical results do not prove the superiority of hierarchical clustering allocation strategies over traditional strategies in risk-adjusted performance. JEL Classification G01, G10, G11 Keywords portfolio selection, hierarchical clustering, HRP, HERC, market stress Title Cluster-based asset allocation strategies during market stress periods
|