-
Comments Off
There is considerable debate among market theoreticians on whether futures prices are random long-term. Fortunately, this debate is not relevant to the analysis of option prices. An option reacts as if the price of its underlying commodity future were a random variable and is not concerned with the direction of the futures market. Recent price direction in a commodity future, then, is irrelevant to the pricing of its options. The size of recent daily price fluctuations in a commodity future, however, is the single most important variable in the pricing of its options.
The value that the free market places on an option is an indication of the price the market expects the commodity future to be trading at the instant the option expires. Even though the most likely outcome is always that the futures price will not have changed at all by the time the option expires, the option market recognizes that there is a range of possibilities for the price of the future, a range of possibilities distributed more or less symmetrically about the unchanged level. Other things being equal, larger expected ranges will result in larger option premiums.
Two variables directly affect the range of possibilities for the price of a future at option expiry. One is the future’s perceived volatility- determined mostly by price patterns of the recent past. The other is time. A commodity future which has been fluctuating a lot in price is more likely to end up with a large cumulative change in price than a commodity future which has been trading in a relatively tight range. And a future with many trading days left till expiry clearly has more opportunity to arrive at an extreme price than one with just a few trading days left.
If daily commodity price changes were true random variables, normally distributed and with mean values of zero, determining the fair value of any commodity option, mathematically, would be possible. Indeed, a massive amount of academic firepower has been directed toward achieving this very goal, on the assumption that futures price changes are normally distributed. The fact that commodity price changes form distributions that are significantly nonnormal renders a great deal of current academic research into option pricing essentially useless, Nobel prizes in economics notwithstanding.