Annualized historical volatility is always calculate with 10-, 20- days time window. I don't quite understand.

  1. Compare with annualized historical return, annualized historical return is never calculated with 10-, 20- days time window. It's simply calculated with a whole year return data. But, in the context of volatility, even a whole year data is at hand, a time window of 10-,20-days is always used.

  2. Compare with moving average, moving average is calculated with a similar time window method. But moving average is not annualized.

Why hist vol is N DAYS ANNUALIZED?

So, the only reason that come to mind is in the context of option trading. N days annualized historical volatility is compared with N days annualized implied volatility in order to find trading opportunities. Is that correct?

  • $\begingroup$ Volatility is relevant over shorter time periods, so usually only the most recent data is used to characterize the assets historical vol. Basically, a stock's volality over the past month is more relevant to any future predictions than using its volatility over the past year or two years. In most models the time units are years, so regardless of the length of the period the volatility data is taken from, it is annualized. $\endgroup$ – Slade Mar 26 '19 at 5:19
  • $\begingroup$ I used to think 1 month hist vol is relevant to 1 month implied vol, which means the length should be identical for historical and future vol. Is that make sense? $\endgroup$ – Shen Mar 26 '19 at 6:53
  • $\begingroup$ @Shen, You may think so. When both statistical volatility and implied volatility are in the state of equilibrium, the option premiums are fairly valued based on historical norms. $\endgroup$ – Dhamnekar Winod Mar 26 '19 at 15:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.