Dynamic Conditional Correlation Multiplicative Error Processes
2016 (English)In: Journal of Empirical Finance, ISSN 0927-5398, E-ISSN 1879-1727, Vol. 36, 41-67 p.Article in journal (Refereed) Published
We introduce a dynamic model for multivariate processes of (non-negative) high-frequency tradingvariables revealing time-varying conditional variances and correlations. Modeling the variables' conditional mean processes using a multiplicative error model, we map the resulting residuals into aGaussian domain using a copula-type transformation. Based on high-frequency volatility, cumulativetrading volumes, trade counts and market depth of various stocks traded at the NYSE, we show thatthe proposed transformation is supported by the data and allows capturing (multivariate) dynamicsin higher order moments. The latter are modeled using a DCC-GARCH specification. We suggest estimating the model by composite maximum likelihood which is sufficientlyflexible to be applicablein high dimensions. Strong empirical evidence for time-varying conditional (co-)variances in tradingprocesses supports the usefulness of the approach. Taking these higher-order dynamics explicitlyinto account significantly improves the goodness-of-fit and out-of-sample forecasts of the multiplicative error model.
Place, publisher, year, edition, pages
2016. Vol. 36, 41-67 p.
Multiplicative error model, Trading processes, Gaussian domain, DCC-GARCH, Liquidity risk
Probability Theory and Statistics Economics and Business
Research subject Statistics
IdentifiersURN: urn:nbn:se:su:diva-127171DOI: 10.1016/j.jempfin.2015.12.002ISI: 000373417300004OAI: oai:DiVA.org:su-127171DiVA: diva2:907193