**How do we make money on Gamma?**

Considering you are long an Option, you make money when realized volatility is greater than the implied volatility & vice versa when you are short an Option.

Is it always true?

1.Answer is No, let’s consider a 1 yr Long call, If realized volatility at the end of the 1 yr is greater than the implied volatility, It’s not necessary that you will make money. Let’s see how.

Daily Pnl for a delta hedged Long call position incase implied vols dont move is : 0.5×Gamma×Spot^{2} {Realized Variance – Implied Variance}

With this expression, we can clearly see that the final P&L is the sum of the daily Variance Spread weighted by the Dollar Gamma. Thus, considering those days when Dollar Gamma is high will tend to dominate the final P&L & Hence it’s the variance spread {Realized Variance – Implied Variance} in those high dollar gamma days that will determine the Pnl, Not the variance spread at the end of 1 yr. For the final P&L to be path-independent, the Dollar Gamma must be constant.

**Intuitive explanation of cross-gamma Pnl:** Assuming we are long an ATM 2-asset worst of put which is near expiry.

Delta of asset 1 @ 99% implied correl: 0.25

Calculation

Case 1: They both move OTM, In that case delta on both is 0

Case 2: They both move ITM, In that case delta on both is 0.5

So expected delta is 0.5×0 + 0.5×0.5 ~ 0.25

Delta of asset 2 @ 99% implied correl: 0.25

Assuming the volatilities of both the assets are similar in magnitude, Now in case the realized correlation is -1, i.e. asset A moves OTM & asset B becomes the WOP, delta on both assets change & to re-balance the delta, we buy 0.75 of B at a lower price since it has gone down & sell 0.25 of asset A (to flatten the delta on A) at a higher price. Hence making money…

Conclusion: (Short cross-gamma ~ Short correlation ~ Short covariance) & Cross gamma Pnl would depend on : **Cross-Gamma *(Realized covariance – Implied Covariance ),** A similar equation with that of Gamma Pnl.

In the above case we are short cross-gamma, i.e. negative cross-gamma & realized covariance(negative) is less than the implied covariance(positive) & hence we make money…

### Like this:

Like Loading...

*Related*

I would expect the crossgamma way the way correl sensi is.

Visually, it is quite intuitive that holding a worst-of, you want asset to decorrelate as much as possible, so that at least one is in in the money.

Comment by nicolas — December 29, 2009 @ 9:58 am |

” Case 1: They both move OTM, In that case delta on both is 0

Case 2: They both move ITM, In that case delta on both is 0.5

So expected delta is 0.5×0 + 0.5×0.5 ~ 0.25″

If the option is near expiry and it’s in-the-money….exactly why would its delta be 0.5?

Comment by Aquarius — January 7, 2010 @ 10:02 am |