EcoSta 2019: Start Registration
View Submission - EcoSta2019
Title: Maximum Lq-likelihood estimation for the mixture of dynamic covariance models Authors:  Lin Xu - Zhejiang University of Finance & Economics (China) [presenting]
Weixin Yao - UC Riverside (United States)
Abstract: A robust estimation is proposed for the mixture of dynamic covariance models based on the maximum Lq-likelihood inference procedure. The model is shown to be identifiable, and can be estimated robustly by a modified EM-type algorithm. Via the constrained quadratic optimization, a within-subject tuning parameter selection criterion is constructed. Additionally, We derive the asymptotic property of the maximum Lq-likelihood estimates. A simulation study and two real data examples are conducted to evaluate the finite sample performance of the proposed methodology.