I used the rugarch package of R v.3.2.5 to fit a hybrid ARMA(1,0)+EGARCH(2,1) model to a series of log-returns on an exchange rate.

My question is:

How should I compute the following measures to describe the conditional variance dynamics of the above hybrid model?

Duration of persistence, ARCH and GARCH effects, Leverage effect, Degree of asymmetry, Intensity of negative shocks, Intensity of positive shocks.

Thank you in advance.

  • $\begingroup$ What have you tried so far? Have you checked the original paper by Nelson that introduces EGARCH? Have you check any textbook, e.g. Tsay "Analysis of Financial Time Series"? $\endgroup$ – Richard Hardy Sep 14 '16 at 16:25
  • $\begingroup$ Yes, I consulted both among other texts. My question is about how to compute those measures for an EGARCH which is not the typical EGARCH(1,1) that is found in many papers. I hope this is understood in my question. $\endgroup$ – msmna93 Sep 14 '16 at 16:39
  • $\begingroup$ You can neglect the ARMA part, then the only difference is EGARCH(2,1) against the standard EGARCH(1,1), right? $\endgroup$ – Richard Hardy Sep 14 '16 at 17:24
  • $\begingroup$ Yes, let's neglect the ARMA part for the sake of this question. $\endgroup$ – msmna93 Sep 14 '16 at 18:44
  • $\begingroup$ It's not a choice to neglect ARMA, it actually does not play a role in your question. So am I right about the last part? $\endgroup$ – Richard Hardy Sep 14 '16 at 19:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.