Template-Type: ReDIF-Paper 1.0
Series: Tinbergen Institute Discussion Papers
Creation-Date: 2022-09-29
Number: 22-070/III
Author-Name: Enzo D'Innocenzo
Author-Workplace-Name: Vrije Universiteit Amsterdam
Author-Name: André Lucas
Author-Workplace-Name: Vrije Universiteit Amsterdam
Title: Dynamic Partial Correlation Models

Abstract: We introduce 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
Classification-JEL: C32, C58
Keywords: Dynamic partial correlations, perturbed stochastic recurrence equations, invertibility, stationarity
File-URL: https://papers.tinbergen.nl/22070.pdf
File-Format: application/pdf
File-Size: 1.531.522 bytes
Handle: RePEc:tin:wpaper:20220070