Activity Number:
|
332
- Multivariate Time Series: Modeling and Estimation
|
Type:
|
Topic Contributed
|
Date/Time:
|
Tuesday, July 30, 2019 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Business and Economic Statistics Section
|
Abstract #306761
|
|
Title:
|
Gaussian Copula Vector Autoregressive Modeling
|
Author(s):
|
Vladas Pipiras* and James Livsey and Benjamin Leinwand
|
Companies:
|
University of North Carolina At Chapel Hill and U.S. Census Bureau and University of North Carolina at Chapel Hill
|
Keywords:
|
Gaussian copula;
vector autoregression;
sparsity;
dynamic factors;
regularization;
principal components
|
Abstract:
|
Multivariate Gaussian copula models driven by sparse or factor vector autoregressions are studied. The focus is on such models of large dimensions and for observations taking integer values. Computationally efficient estimation methods based on covariances and their relationships involving Hermite expansions are introduced and examined, both numerically on synthetic and real data, and from a theoretical standpoint.
|
Authors who are presenting talks have a * after their name.