On the Economic Determinants of Optimal Stock-Bond Portfolios: International Evidence
Updated version of: University of Heidelberg, Department of Economics, Discussion Paper Series No. 636
54 Pages Posted: 19 Jul 2017 Last revised: 11 Dec 2017
Date Written: November 6, 2017
Abstract
Using a modified DCC-MIDAS specification that allows the long-term correlation component to be a function of multiple explanatory variables, we show that the stock-bond correlation in the US, the UK, Germany, France, and Italy is mainly driven by inflation and interest rate expectations as well as a flight-to-safety during times of stress in financial markets. Based on covariance forecasts from the new specification, we construct stock-bond hedge portfolios and show that these portfolios outperform various benchmark portfolios in terms of portfolio risk. While optimal daily weights minimize portfolio risk, we find that portfolio turnover and trading costs can be substantially reduced when switching to optimal monthly weights.
Keywords: stock-bond correlation, DCC, DCC-MIDAS, survey data, macro expectations, forecasting, portfolio choice, asset allocation
JEL Classification: C32, C58, E44, E52, G11, G15, G17
Suggested Citation: Suggested Citation