Vine copula base modelling and forecasting of multivariate realized volatility time-series

• Applicants: Professorin Dr. Claudia Czado; Professor Dr. Yarema Okhrin
• Subject Area: Statistics and Econometrics
• Term: from 2015 to 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 263890942

• Final Report Year: 2021

Final Report Abstract

This project contributed to development of vine copula based models under different data structures occurring in econometrics and life sciences. These improved estimation and forecasting of realized volatility time series over standard approaches and as well provided first time development of vine based dependence models under right-censoring.

Publications

• (2018) Vine copula based likelihood estimation of dependence patterns in multivariate event time data. Computational Statistics & Data Analysis 117: 109-127
  Barthel, Nicole, Candida Geerdens, Matthias Killiches, Paul Janssen, and Claudia Czado
  (See online at https://doi.org/10.1016/j.csda.2017.07.010)
• (2019) Dependence modeling for recurrent event times subject to right-censoring with D-vine copulas. Biometrics 75.2: 439-451
  Barthel, Nicole, Candida Geerdens, Claudia Czado and Paul Janssen
  (See online at https://doi.org/10.1111/biom.13014)
• (2019) Modelling temporal dependence of realized variances with vines. Econometrics and Statistics 12: 198-216
  Czado, Claudia, Eugen Ivanov, and Yarema Okhrin
  (See online at https://doi.org/10.1016/j.ecosta.2019.03.003)
• (2020) A partial correlation vine based approach for modeling and forecasting multivariate volatility time-series. Computational Statistics & Data Analysis 142
  Barthel, Nicole, Claudia Czado, and Yarema Okhrin
  (See online at https://doi.org/10.1016/j.csda.2019.106810)