Dear
Prof.,
I am very sorry to
disturb you for my question. I am a PhD student in Electrical Engineering of
Strathclyde University. It is no doubted that you are an expert on the GARCH
modelling. In my research, I used GARCH(1,1) model to evaluate the volatility on
electricity price in the deregulated power market because the time series of
electricity price consist of fat tail and volatility clustering characteristic.
However, when using the GARCH model to evaluate the price volatility, it needs
to compare the modelled volatility values and the real ones? That is, it would
be convenient to include some numerical quantification of the accuracy of the
GARCH method: mean error, maximum error, and residual distribution? How to
define the real volatility?
My second
question is :
In general,
the volatility of electricity price is defined as the rate of electricity price
returns. However, the definition of volatility on GARCH model is
referred as the next period variance? What is the difference?
I am very
sorry that I am not a student of financial department. However, this topic will
play an important role in the future power market.
Thanks you
very much
Best
regards,