A new working paper from that I co-authored together with André Lucas is now available as Tinbergen Institute Discussion Paper.
You can take a look at the full paper here.
The paper introduces a new, easily scalable model for dynamic conditional correlation matrices based on a recursion of dynamic bivariate partial correlation models.
By exploiting the model’s recursive structure and the theory of perturbed stochastic recurrence equations, we establish stationarity, ergodicity, and filter invertibility in the multivariate setting using conditions for bivariate slices of the data only.
From this, we establish consistency and asymptotic normality of the maximum likelihood estimator for the model’s static parameters.
The new model outperforms benchmarks like the t-cDCC and the multivariate t-GAS, both in simulations and in an in-sample and out-of-sample asset pricing application to 1980–2021 US stock returns across twelve industries.
You can also take a look at the Poster or download the paper presentation, but if you have further questions feel free contact me. If you have some comments, please leave them below.