Contents
- 1 Is there a method to find the optimal lag for the GARCH model?
- 2 Which is the estimator of the optimal lag length?
- 3 How is the BIC estimate related to the lag order?
- 4 How to test normality of residuals in GARCH model?
- 5 Do the other studies assume the same conditional mean and conditional variance dynamics?
Is there a method to find the optimal lag for the GARCH model?
Is there a method to find the optimal lag for the GARCH model? Edit: I used “fGarch” package in R to fit a GARCH (1,1) model. Here is the output:
Which is the estimator of the optimal lag length?
Both criteria are estimators of the optimal lag length p p. The lag order ˆp p ^ that minimizes the respective criterion is called the BIC estimate or the AIC estimate of the optimal model order.
The lag order ˆp p ^ that minimizes the respective criterion is called the BIC estimate or the AIC estimate of the optimal model order. The basic idea of both criteria is that the SSR S S R decreases as additional lags are added to the model such that the first term decreases whereas the second increases as the lag order grows.
How to test the significance of a lag?
Estimate an AR ( p p) model and test the significance of the largest lag (s). If the test rejects, drop the respective lag (s) from the model. This approach has the tendency to produce models where the order is too large: in a significance test we always face the risk of rejecting a true null hypothesis!
Can a GARCH model be combined with an arch model?
As we have seen, an AR(1) process has a nonconstant conditional mean but a constant conditional variance, while an ARCH(1) process is just the opposite. If both the conditional mean and variance of the data depend on the past, then we can combine the two models. model with any of the GARCH models in Section 18.6.
How to test normality of residuals in GARCH model?
All coefficients are significant. p -values of Jarque-Bera test and ARCH-LM test are greater than 0.05. Can I use this as a good model? How can I test normality of residuals?
Do the other studies assume the same conditional mean and conditional variance dynamics?
Do the other studies also assume the same conditional mean and conditional variance dynamics of the univarate models, i.e. ARMA (1,1)-GARCH (1,1)? I suppose ARMA (1,1) could be omitted in favour of just a constant. [Apparently this does not generate the desired result.] What if you fit the model fo subsamples of your data?