You might try looking at Carol Alexander's book "Market
Models" as an introduction for GARCH models and what
to do with them.
Patrick Burns
Burns Statistics
patrick@burns-stat.com
+44 (0)20 8525 0696
http://www.burns-stat.com
(home of S Poetry and "A Guide for the Unwilling S User")
Wu Yuan-Kang wrote:
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,
---------------------------------------------------------------------------------
Wu Yuan-Kang
Power Systems Research Group
Department of Electronic and Electrical Engineering
University of Strathclyde
204 George Street, Glasgow G1 1XW, United Kingdom
Tel: 44 141 5482638
|