The above graph is a visual representation of the Black-Scholes model. Or Black-Scholes-Merton if you want credit shared equitably and "you're not into the whole brevity thing," Mr Lebowski.
As you can see, there are three
axes. The x axis (the width of the box) is the price of the underlying asset,
the stock price, treating the strike
price in the center as 1. The y axis (the height of the box) is the volatility
of that option. The z axis (the depth of the box) is the time to maturity [and
either exercise or expiration].
As you can see, the plane of various
shades of blue (the darker the blue, the higher) has a sharp crease at the
front/center/bottom of the box, where the sharp difference between winners and
losers on the lottery’s drawing date is indicated.
We should also mention that the
volatility that goes into the calculations as we’ve described them above is historical volatility. One of the
assumptions of the BSM model is constant volatility. This is a counter-factual
assumption: volatility does in fact change.
Another phrase you might want to
remember is: implied volatility. That comes up when a theorist is doing his
calculations in the other direction. If we know the present price of both the
underlying stock and the stock option, what degree of historical/constant
volatility does their relationship imply?
As you might have grasped by now,
the elegance of the BSM model, the elegance of that neatly sloping blue plane
in the box, comes at some cost in heroic assumptions. We have to assume that
the movement of stock values describes a bell curve rather than a non-normal
sort of curve. We also have to assume that the standard deviation of that bell
curve remains constant over time. We also have to assume the degree of
liquidity and transparency in the markets necessary to make arbitrage easy,
because that assumption is behind the various proofs of these equations.
Does all that make the model
worthless? Not at all. As Emanuel Derman has written, all models sweep dirt
under the rug. A good model is one that “makes explicit the dirt swept away,”
and BSM is decidedly a good model in this sense.
BSM was a beachhead of an invasion.
In showed the way, and other intellectual troops rushed in to widen the
territory covered. Stock options, after all, are just one example of a broad
category of instrument, known as derivatives. The value of a stock option
depends upon that of the underlying stock, and in like manner the value of
other derivatives depends upon the value of other sorts of underlying asset.
The brain power of the quantitative
analysts (“quants”) of the world focused on this point: how far could the
underlying logic of the BSM model be extended to other sorts of derivative? The
answer turns out to be: quite far indeed. Yet extending its realm doesn’t
change the nature of the assumptions built into it, nor the fact of their
fallibility.
Comments
Post a Comment