After the estimation of volatility using the GARCH (1,1) model, to call it a good model we need to run the residual diagnostic tests namely
1. Correlogram - squared residual test
2. ARCH test.
3. Normal distribution test
The question here is:
when we run Correlogram - squared residual test How many lags do we need to consider?
John Alexander Pulgarin Diaz
Suresh, the number of lags to consider is depending on your analysis. Remember that we expect 5% of the estimates to present autocorrelation, using an alfa=0.05.
This book could be of some help:
- Metcalfe, A. V., & Cowpertwait, P. S. (2009). Introductory time series with R.
- Dale, M. R., & Fortin, M. J. (2014). Spatial analysis: a guide for ecologists. Cambridge University Press.
Best regards!
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publications
More Suresh A S's questions See All
Similar questions and discussions