Extending the variance and measurement equation in GARCH models in the rugarch package (Realized GARCH)
This post was updated on .
I am wondering if you are able to give me some direction. I am currently writing my thesis and have been using the rugarch package for the GARCH modelling.
I am using the realized GARCH model, and I want to extend the realized GARCH variance equation by including two parameters. These parameters are: the lagged 5 day average RV measure and the lagged 22 day average RV measure. The measurement equation remains the same, and thus the likelihood functions should be the same.
The variance equation would be: log(σ2t) =ω+∑γ1log(RVt−1) +∑γ2log(RVt−2) +∑βlog(σ2t−1) + ∑θ1log(5davgRVt-1) + ∑θ2log(22davgRVt-1)
This is similar to a paper by Huang, Liu and Wang, named Modeling long memory volatility using realized measures of volatility: A realized HAR GARCH model. Huang is one of the authors of the original Realized GARCH model.
EDIT: I figured that I can do this with the external.regressors function. Having done some estimations and forecasting, the results are negligible in comparison to the original realized GARCH specification, contrary to the aforementioned paper. I assume this is because they have extended the measurement equation too. Is this possible with the rugarch package?
Thank you for your time,
[[alternative HTML version deleted]]
R-SIG-Finance@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance -- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should go.