I have take index return for 20 years and extracted a part of series based upon certain timing (similar to seasonal effect). Now i want to analyse if there is high volatility during the segregated series and the spillover.
Here the issue is to check how your cross correlations and squared correlations behave.
You must ensure that your modeling meets the basic assumptions of the model, that is, that it is stationary.
After you have stationary models you must choose the best one among them under a statistical selection criteria like the AIC. In the outputs you share the first is better than the second.
After you check that the models are stationary, you make the prediction.
This is a print out generated by software with your data input. It runs the basic test for the model you specified. If you have problem reading the pring out, here are some materials on basic GARCH. Mastter basic concept of GARCH would help you read the print out quite easily.
J. Marcelo Miranda, I had considered the basic assumptions of the model before proceeding with the analysis, all I need to check is a volatility in a given time series (a) is higher than the other (b), conditioning** (a & b) are sub-series of a single series (c). So to the extend i found was, garch family models can give the best result especially EGarch and GJR Garch.
I am facing troubles in interpretation of the results. Thanks for your suggestions to verify the assumptions before proceeding with model selection
Paul Louangrath, you got it right sir, i am facing problem in interpreting the results accurately. Thanks for the guidance material