My data set is a time series for daily returns for a stock market index, spanning over 10 years Before applying a GARCH model, what are the requirements my data set should have and what are the tests I should run?
Mainly, volatility clustering and asymmetric volatility, found often in log-return series of financial and commodity prices.
Volatility clustering, large price changes tend to be followed by large changes, but random sign.
Asymmetric volatility, large declines in prices tend to be followed by larger and more persistent price volatility than equal magnitude increases in prices.
For recent extension of standard GARCH models, see
Savva, C and P. Theodossiou, 2018. “The Risk and Return Conundrum Explained: International Evidence,” Journal of Financial Econometrics, June 2018, Vol. 16, No. 3, pp. 486 – 521. https://doi.org/10.1093/jjfinec/nby014
Theodossiou, P and C Savva, 2016. “Skewness and the Relation between Risk and Return,” Management Science, 2016, Vol. 62, No. 6, 1598–1609.
The data must be Stationary, if not you can differentiate it.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics