摘要：The aim of this paper is to examine the impact of time-varying distributional parameters on portfolio performance and risk. The main findings show that the Exponentially Weighted Moving Average (EWMA) model provides the lowest level of risk for efficient portfolios. Constant parameters Markowitz and DCC-GARCH (1,1) models producing similar results, both inferior to the EWMA procedure. Rolling EWMA efficient portfolios indicate how the portfolio weights should be amended over time. This result is important at times of the turbulence in the financial markets as it may help to protect the portfolio against big losses.
