Abstract
A vector error correction model is proposed for forecasting realized volatility which takes advantage of the cointegration relation between realized volatility and implied volatility. The model is constructed by adding a cointegration error term to a vector-and-unit-root version of the heterogeneous autoregressive (HAR) model of Corsi (2009). The proposed model is easier to implement, extend, and interpret than fractional cointegration models. A Monte Carlo simulation and real data analysis reveal advantages of the proposed model over other existing models of Corsi (2009), Busch Christensen and Nielsen (2011), Cho and Shin (2016), and Bollerslev Patton, and Quaedvlieg (2016).
| Original language | English |
|---|---|
| Pages (from-to) | 1503-1515 |
| Number of pages | 13 |
| Journal | Communications in Statistics: Simulation and Computation |
| Volume | 48 |
| Issue number | 5 |
| DOIs | |
| State | Published - 28 May 2019 |
Bibliographical note
Funding Information:The authors are very grateful for the constructive comments of two referees which substantially improve the paper. This study was supported by a grant from the National Research Foundation of Korea (2016R1A2B4008780)
Publisher Copyright:
© 2017, © 2017 Taylor & Francis Group, LLC.
Keywords
- Cointegration
- HAR model
- High frequency data
- Long-memory
- Volatility forecasting