Abstract
ABSTRACTThis article uses the GARCH-MIDAS model to decompose Taiwan stock volatility and studies the role of US economic variables in each component. The full-sample results indicate that the additional explanatory information of US variables is contributed mostly by stock market measures, and the link between short-run Taiwan and US stock volatility is particularly evident. The out-of-sample results suggest that the in-sample significant US variables lead to slightly smaller forecast errors for both volatility components, but the improvements are very limited. The analysis also extends to Electronics and Non-Electronics subindices, a range-based volatility estimator and a different volatility decomposition method. Despite these alternatives, the main conclusions do not change.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call