Affordable Access

Publisher Website

Long memory and nonlinearity in conditional variances: A smooth transition FIGARCH model

Authors
Journal
Journal of Empirical Finance
0927-5398
Publisher
Elsevier
Publication Date
Volume
18
Issue
2
Identifiers
DOI: 10.1016/j.jempfin.2010.11.007
Keywords
  • Figarch
  • Stfigarch
  • Volatility
  • Long Memory
  • Smooth Transition
  • Asymmetry
Disciplines
  • Design

Abstract

Abstract This paper introduces the Smooth Transition version of FIGARCH model which is designed to account for both long memory and nonlinear dynamics in the conditional variance. Nonlinearity is introduced via a logistic transition function. The model can capture smooth changes in the volatility across different regimes as well as asymmetric response to negative and positive shocks and allows for nonzero thresholds. Simulations find that the Smooth Transition FIGARCH model outperforms the standard FIGARCH model when nonlinearity is present, and ignoring nonlinearity in the data may induce considerable costs in terms of bias and efficiency. Applications to exchange rate and stock market data show that the proposed model performs well both in-sample fit as well as in forecasting one-day ahead volatility.

There are no comments yet on this publication. Be the first to share your thoughts.