This note discusses some aspects of the paper by Hu and Tsay (2014), “Principal Volatility Component Analysis”. The key issues are considered, and are also related to existing conditional covariance and correlation models. Some caveats are given about multivariate models of time-varying conditional covariance and correlation models.
# 14-025/III (2014-02-25)
- Michael McAleer, College of Technology Management, National Tsing Hua University, Taiwan; Econometric Institute, Erasmus University Rotterdam, the Netherlands; Complutense University of Madrid, Spain
- Principal Component Analysis, Principal Volatility Component Analysis, Vector time-varying conditional heteroskedasticity, BEKK, DCC, asymptotic properties
- JEL codes:
- C32, C55, C58, F37