|
Activity Number:
|
582
- Nonparametric Methods for Statistical Inference
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
|
Abstract #304103
|
|
|
Title:
|
Proxy Variables to Common Factors and Parameter Estimation in Factor Copula Models
|
|
Author(s):
|
Pavel Krupskiy* and Harry Joe
|
|
Companies:
|
University of Melbourne and University of British Columbia
|
|
Keywords:
|
conditional independence;
factor copula;
maximum likelihood estimation;
tail dependence;
tail asymmetry
|
|
Abstract:
|
Factor copula models assume observed variables are independent conditional on one or several unobserved factors. We show that, under some mild assumptions, proxy variables to the unobserved factors can be obtained from the observed variables. These proxy variables can help to select appropriate linking copulas in factor copula models and to get fast estimates of parameters of these copulas in high dimensions. We use simulation study to show that parameter estimates obtained using the proposed approach are very close to estimates obtained using the maximum likelihood approach which can be computationally demanding for big data sets. We apply the proposed approach to analyze a financial data set.
|
Authors who are presenting talks have a * after their name.
