wilmott.com

Hi,Could someone explain as to why do we calculate volatility of an asset as stdev of logn of the returns.As in, why do we take log of the returns to calculate vols. Basically I wanted to know the intuition behind the same.Suppose we calculated vols by simply taking the stdev of a series, would it be correct in option pricing.

The vol in the Black Scholes model is the standard deviation of the continuously compounded return of the underlying. Say you observe two sucessive price observations of the underlying P0 and P1. The continuously compounded return is defined by P1 = P0*exp(R*T). Solve for R.

outrun...nice! this is a picture i have not seen in probably 20 years or something? you made my day i finally gotta get myself one of those C64 or Amiga emulators.

Sorry. Duplicate post.

QuoteOriginally posted by: STJCould someone explain as to why do we calculate volatility of an asset as stdev of logn of the returns. As in, why do we take log of the returns to calculate vols. Basically I wanted to know the intuition behind the same.Because it is an estimate of volatility based on the assumption of a lognormal distribution. We take log returns to check how well the returns correspond to a lognormal distribution. We conjecture that a lognormal distribution is a better decsription than a normal distribution because the asset price is never negative and we expect the scale of variation to be roughly proportional to the price (other things being equal)QuoteSuppose we calculated vols by simply taking the stdev of a series, would it be correct in option pricing.No. and neither is it for log returns either. The standard deviation of log returns over a given historic time period, the instantaneous volatility and the volatility used for pricing an option over a future time to expiration are, in general, three different quantities. They become the same if the instantaneous volatility is constant and the time step of the time series shrinks to zero (or the number of steps is infinite or the drift rate vanishes).