1
$\begingroup$

On Page 24 of N. Taleb's "Dynamic Hedging" the author gives the following example

Example: Assume that an asset trades at \$100, with interest rates at 6% (annualized) and volatility at 15.7%. Assume also that the 3-month 80 call is worth \$20, at least if it is American. Forgoing early exercise would create an opportunity cost of 20 x 90/360 x .06 = .30 cents, the financing of \$20 premium for 3 months. The time value of the equivalent put is close to zero (by put-call parity), so the intelligent operator can swap the call into the underlying asset and buy the put to replicate the same initial structure at a better cost. He would end up long the put and long the underlying asset.

The possible position of the operator before swap:

  • 1 call worth of \$20 a
  • \$80 in cash.

The position after the swap:

  • 1 asset worth of $100 at the current spot price
  • 1 put worth of almost zero

If I could earn 6% both on \$80 in cash and on 1 asset (i.e. if the asset is another currency for example) and the spot price would remain the same \$100 then I would agree with the calculations of the author:

1 asset x 90/360 x .06 - \$80 x 90/360 x .06 = (100 - 80) x 90/360 x .06 = 30 cents

But if the price of the asset will go down to say \$75, then I'd better stay with the call because:

1 asset x 90/360 x .06 - \$80 x 90/360 x .06 = (75 - 80) x 90/360 x .06 = -7.5 cents

So,how far the spot price is likely to go from the current level in three months if its volatility is 15.7%?

$\endgroup$
1
$\begingroup$

Keep in mind that there is nothing dynamic about this at all...it is only a snapshot and it is only a 1 sigma range.

High side:

([Price] * (1 + ([Vol] * SQRT[days to expiry]/365)))
(100 * (1 + (.157 * sqrt(90/365)))
107.7960

Low side:

([Price] * (1 - ([Vol] * SQRT[days to expiry]/365)))
(100 * (1 - (.157 * sqrt(90/365)))
92.2040
$\endgroup$
  • $\begingroup$ So Taleb is right - under the assumptions the spot price is unlikely to go below 80, so an intelligent operator should swap $\endgroup$ – zer0hedge Mar 15 '17 at 6:52
  • 1
    $\begingroup$ Depends on how you quantify the term "likely". $\endgroup$ – Quantuple Mar 15 '17 at 8:00
  • $\begingroup$ @Quantuple In this example, the price will go below 84 with less than 5% probability, which might be considered unlikely $\endgroup$ – zer0hedge Mar 20 '17 at 18:39
  • $\begingroup$ @zer0hedge it's actually even smaller than that should you consider a BS world where log-returns are normally distributed. But this does not change my statement :) $\endgroup$ – Quantuple Mar 20 '17 at 18:52

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.