Choosing Volatility Inputs for Bond Options in the Black 76 Model
Summary
The document considers whether recent historical bond-price volatility is suitable as the volatility input for pricing a bond option with the Black 76 model. The preferred input, when available, is implied volatility: the value that reproduces the option’s current market price when used in the model. For bond options without reliable implied volatility, historical volatility is offered as an estimate.
The replies caution that the estimate depends on the observation window. A 30-day lookback is only one choice; matching the lookback to the option’s maturity is suggested as a possible approach, though not a universal rule. For thinly traded bond options, implied volatilities from liquid related instruments, such as options on Treasury futures, may provide a useful reference or surface. Historical volatility is backward-looking, so its use is most defensible when market-implied data is unavailable, unstable, or unsuitable for the task.
Key ideas
- Use market-implied volatility when a reliable option price is available.
- Historical volatility can estimate the input when implied volatility is missing or unsuitable.
- The historical lookback window affects the estimated volatility and resulting option value.
- Volatility from liquid options on related underlyings can inform pricing of less liquid bond options.
- Historical volatility is backward-looking and may not reflect forward market expectations.
Tags
Full text
# Use of Historical Volatility in Black 76 Model # Use of Historical Volatility in Black 76 Model I am trying to use the Black 76 model to calculate the price of a bond option. Is it possible to use the historical volatility of the bond prices (say standard deviation of the log returns over the last 30 days) as the volatility input into the Black 76 formula? If not, then what should I be using as the volatility input? Any help would be greatly appreciated. Thanks! ## Answer by Juan Ignacio Gil (score 2, accepted) https://quant.stackexchange.com/a/37342 Ideally, you should use the implied volatility: this is the volatility that, when input into the Black formula, returns the price that the option has right now in the market. But if you don't have this data (sometimes, depending on purpose, even if you had this data), you have to get an estimation of the volatility. In this case, using the historical volatility is a good choice. Keep in mind that using 30 days is just a choice, and that you may get different results (so different option prices) if you use shorter or longer periods (I tend to use as a period the time to maturity: for an option with a 6 month maturity I would use 6 months of data, but, again, this is a disputable choice) ## Answer by League Super (score 0) https://quant.stackexchange.com/a/75148 You should try to start with implied volatilities as long as you have other financial instruments on similar underlyings that have good liquidity and can be priced using Black 76. In this case, let's say you want to price treasury bond options. These are most likely OTC traded. Assume there are abundant trades on treasury futures options(related to bonds) traded on exchanges. Then it is a good idea to get a volatility surface from treasury futures options of different maturities and use these vols as input parameters. Historical vols are backward-looking and your options are forward-looking, so it would be more natural to use implied vols in general. However in markets where products are thinly traded and implied vols are not stable or even available, you can use historical vols as long as it justifies your goals.
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.