|
Activity Number:
|
92
- Time Series and Finance
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 9, 2021 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
Business and Economic Statistics Section
|
|
Abstract #318707
|
|
|
Title:
|
Estimating Factors in Dynamic Equicorrelation Model
|
|
Author(s):
|
Raja Velu* and Zhaoque Zhou
|
|
Companies:
|
Syracuse University and Syracuse University
|
|
Keywords:
|
Large Covariance Matrix;
Equicorrelation;
Factor Models;
Hetereoscadsticity;
Asset Pricing
|
|
Abstract:
|
Engle and Kelly (2012) propose a covariance matrix estimator with the assumption that all pairwise correlations are equal. This assumption holds in various areas of finance and can be used to model large covariance matrices that are time varying. In this paper, we construct factors under this equicorrelation model and provide a reasonable and easily computed approximation of the largest eigenvalue of the covariance matrix and provide analytical structure of the factor loadings. This helps to monitor the dependence of the heteroscedastic returns over time. Our results can be also applied in portfolio construction and in asset pricing.
|
Authors who are presenting talks have a * after their name.
