I've found that very little in financial analysis is precise (including that square-root-of-time stuff), but I suggest that one use whatever definition happens to be in vogue ... so everybuddy is speaking the same language However, I'd be interested in how you would calculate annual volatility?
It depends for what purpose. Trading ? I would rather use ATR. Option pricing ? sorry, my methods are only for sale Portfolio management ? sorry again, the method is for sale
That's life... my methods (but I'm not the one to sell them as I got paid to develop them for others) are for sale, but you make money out of them. The standard method is free...but you don't get a penny...and usually get fucked by people using my (or others) method...
I've noted (with amusement) science_trader's cryptic statements: "It's just a question of Hurst exponent."** "something totally inaccessible to people not trained in probs" and even the obvious "if the market is an infinite-variance process, 'volatility' is infinity" ** Actually, I'd explain why the Hurst exponent has nothing whatsoever to do with skew or kurtosis, but my explanation is for sale ...
Did I say that Hurst exponent was in relation with skew or kurthosis ? I don't think so. Next time I'll write a complete sentence for people to understand better.
No, you didn't, but I'd still be happy to prove* that it doesn't ... for a price *There's this thing I have about mathematical "proofs" of events that occur in the real world ... in particular, "proofs" of things financial. One of my favourite quotes: "As far as the laws of mathematics refer to reality, they are not certain, and as far as they are certain, they do not refer to reality." ... Albert Einstein
Beware of the multifractality of prices in your study Don't give Einstein to much credit for his works. I have studied physics and I didn't find his works were really so great (with all due respect). I much better appreciate Feynman's ideas...and life