Each week, we will be publishing layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.

The article featured today is from Applied Stochastic Models in Business and Industry, with the full article now available to read in Early View here.

Pape, K, Galeano, P, Wied, D. Sequential detection of parameter changes in dynamic conditional correlation models. Appl Stochastic Models Bus Ind. 2020; 1– 21. https://doi.org/10.1002/asmb.2578

A multivariate monitoring procedure is presented to detect changes in the parameter vector of the Dynamic Conditional Correlation (DCC) model originally proposed by R. Engle. The procedure can be used to detect changes in both the conditional and unconditional variances as well as in the correlation structure of the model. Such information could principally be used for constructing optimal portfolios or to anticipate crises since volatilities and correlations tend to increase in turbulent market phases. While former monitoring methods for multiple asset returns often focus either on variances or correlations, we aim at monitoring structural changes in both volatilities and correlations jointly.

The detector is based on the contributions of individual observations to the gradient of the quasi-log-likelihood function. More precisely, standardized derivatives of quasilog-likelihood contributions at time points in the monitoring period are evaluated at parameter estimates calculated from a historical period in which there are no changes. The null hypothesis of a constant parameter vector is rejected if these standardized terms differ too much from zero, which is the expected value if there have been no changes. Critical values needed to apply the test are obtained via a parametric bootstrap type procedure. Size and power properties of the procedure are examined in a simulation study. Finally, the behavior of the proposed monitoring scheme is illustrated with a group of asset returns of several insurance companies. The procedure allows the time series of asset returns to be split up into calm and more turbulent phases linked to financial and debt crises.