0
$\begingroup$

Modelling financial time series data.

My mean equation was an AR(1) process. Planning to forecast the data in-sample and I have a TGARCH with IC indicating it is a better model, but the p-value for the AR(1) is .0566, whereas an EGARCH model has a significant (5%) AR(1) value but its IC indicate the TGARCH is superior.

Which should I use? (Worth mentioning the aim is to have the best fit model)

$\endgroup$
  • $\begingroup$ The use of statistical significance vs. information criteria for model selection has been discussed in multiple previous posts. You could benefit from checking them out. $\endgroup$ – Richard Hardy Apr 7 '18 at 15:47
  • $\begingroup$ Any links? Have looked briefly but unable to find anything concrete. Would really appreciate anything. $\endgroup$ – BAMIR Apr 8 '18 at 17:39

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.