Hi zhihui，
Yes, what you say is true. There is no definite reason to define volatility on annual basis.
However if you input daily return without annual adjustments, you should be aware that your output vol is also annualized at all.
Regards,
Cheng
发件人: 钱哲辉 [mailto:[hidden email]]
发送时间: 2014年6月23日 15:58
收件人: cheng.li; quantlib-users
主题: Re: 答复: [Quantlib-users] Calibration of GARCH11 Model in QuantLib
Hi Cheng,
Thank you for your answer! However, I'm quite sure there is volatility on daily basis and this kind of volatility appears in many cases. For example, we can put daily returns, like ln(S2/S1), into the GARCH model in Matlab to calibrate the parameters. Thanks again.
Regards,
Zhehui
发件人： [hidden email]
发送时间： 2014-06-23 15:32
收件人： [hidden email]; [hidden email]
主题： 答复: [Quantlib-users] Calibration of GARCH11 Model in QuantLib
That means the input yield should be annualized. Otherwise the output is not volatility at all (when we talk about volatility it is always on annual basis)
Regards,
Cheng
发件人: 钱哲辉 [[hidden email]]
发送时间: 2014年6月20日 15:27
收件人: quantlib-users
主题: [Quantlib-users] Calibration of GARCH11 Model in QuantLib
Hi,
I am new in using QuantLib. Suppose I have the everyday close prices of S&P 500 index, next I have to convert these prices to yield rates. In common sense, daily yield is needed to calibrate GARCH11 Model. However, the comments in Garch11(a class) of QuantLib says, 'Volatilities are assumed to be expressed on an annual basis'. And I get confused. Which kind of yield should I put into the constructor of Garch11 in QuantLib? Hope someone could help me out, a simple piece of code is appreciated.
Thanks.
Zhehui Qian
| Free forum by Nabble | Edit this page |