CFE 2020: Start Registration
View Submission - CMStatistics
Title: Time-heterogeneous D-vine copula model for longitudinal data Authors:  Md Erfanul Hoque - University of Manitoba (Canada) [presenting]
Elif Acar - University of Manitoba (Canada)
Mahmoud Torabi - University of Manitoba (Canada)
Abstract: Longitudinal studies collect repeated measurements from subjects over time to understand the dependence mechanisms among these measurements. However, in many cases, the number and timing of measurements differ across study subjects so that the data may be unbalanced and unequally spaced and may have an impact on the dependence structure of such data. Hence, statistical analysis of such data requires accounting for both the unbalanced study design and the spacing of repeated measurements. We propose a time-heterogeneous D-vine copula model that allows for time adjustment in the dependence structure of unequally spaced and potentially unbalanced longitudinal data. Moreover, we investigate the asymptotic properties of the parameter estimates under proposed models. The performance of the time-heterogeneous D-vine copula models is evaluated through simulation studies and by a real data application.