Linkages between asset classes during the financial crisis, accounting for market microstructure noise and non-synchronous trading

In this paper we analyse market co-movements during the global financial crisis.  Using high frequency data and accounting for market microstructure noise and non-synchronous trading, interdependencies between differing asset classes such as equity, FX, fixed income, commodity and energy securities are quantified.  To this end multivariate realised kernels and GARCH models are employed.  We find that during the current period of market dislocations and times of increased risk aversion, assets have become more correlated when applying these intra-day measures.  FX pairs seemingly lead the other variables, but commodities remain entirely unaffected.