Deriving Black–Scholes Call Vega from the Strike and Spot Terms
Summary
The document checks whether differentiating the Black–Scholes call price with respect to volatility produces the familiar vega formula. The derivation starts from the two density-weighted terms involving the spot price and discounted strike, then factors out the standard normal density at d-plus. It uses the relationship between d-plus and d-minus to simplify the exponential ratio and combine the remaining terms.
The result is call vega equal to spot multiplied by the standard normal density at d-plus and the square root of time to expiry. An equivalent expression uses the discounted strike and the density at d-minus. The explanation assumes the standard Black–Scholes setup and the stated definitions of d-plus and d-minus; the document does not discuss dividends, alternative volatility conventions, or other model extensions.
Key ideas
- Call vega is the derivative of the Black–Scholes call price with respect to volatility.
- Factoring out the standard normal density at d-plus helps simplify the differentiated price expression.
- The exponential ratio links the d-plus and d-minus density terms to spot and discounted strike.
- Vega can be written using either spot with the density at d-plus or discounted strike with the density at d-minus.
- The resulting formula includes the square root of time to expiry.
Tags
Full text
# Derive vega for Black-Scholes call from this formula?
# Derive vega for Black-Scholes call from this formula?
Is it possible to get the right formula for vega of a call option under the black scholes model from this formula?
$$\frac{\partial{C}}{\partial{\sigma}}=\frac{S_0}{\sqrt{2\pi}}{e^\frac{-d_+^2}{2}}(\frac{-1}{\sigma})(d_-)-\frac{Ke^{-rt}}{\sqrt{2\pi}}e^{\frac{-d_-^2}{2}}(\frac{-1}{\sigma})(d_+)$$
$d_-=\frac{\ln{\frac{S_0}{k}}+(r-\frac{\sigma^2}{2})t}{\sigma\sqrt{t}}$ $d_+=\frac{\ln{\frac{S_0}{k}}+(r+\frac{\sigma^2}{2})t}{\sigma\sqrt{t}}$
## Answer by Gordon (score 9, accepted)
https://quant.stackexchange.com/a/26155
Note that, \begin{align*} \frac{\partial{C}}{\partial{\sigma}} &=\frac{S_0}{\sqrt{2\pi}}{e^\frac{-d_+^2}{2}}(\frac{-1}{\sigma})(d_-)-\frac{Ke^{-rt}}{\sqrt{2\pi}}e^{\frac{-d_-^2}{2}}(\frac{-1}{\sigma})(d_+)\\ &=\frac{1}{\sqrt{2\pi}}e^{\frac{-d_+^2}{2}}\left[-\frac{S_0 d_-}{\sigma} + \frac{Ke^{-rt}d_+}{\sigma} e^{\frac{d_+^2}{2} - \frac{d_-^2}{2}} \right]\\ &=N'(d_+)\left[-\frac{S_0 d_-}{\sigma} + \frac{Ke^{-rt}d_+}{\sigma} e^{\frac{1}{2}(d_+-d_-)(d_++d_-)} \right]\\ &=N'(d_+)\left[-\frac{S_0 d_-}{\sigma} + \frac{Ke^{-rt}d_+}{\sigma} e^{\frac{1}{2}\sigma \sqrt{t}\, \frac{2\ln \frac{S_0}{K} +2rt}{\sigma \sqrt{t}}} \right]\\ &=N'(d_+)\left[-\frac{S_0 d_-}{\sigma} + \frac{S_0d_+}{\sigma} \right]\\ &=S_0 N'(d_+)\sqrt{t}, \end{align*} which is the Black-Scholes vega formula.
## Answer by jaehyukchoi49 (score 4)
https://quant.stackexchange.com/a/71139
The answer by @Gordon is pretty complete, but let me add one more point. Let $n(x) = N'(x)$ be the PDF of standard normal distribution.
In the derivation, note that $$ e^{d_+^2/2 - d_-^2/2} = \frac{n(d_-)}{n(d_+)} = \frac{S_0}{Ke^{-rt}}. $$ Thanks to this relation, there are two equivalent expressions for the Black-Scholes vega: $$ \frac{\partial C}{\partial \sigma} = S_0 n(d_+) \sqrt{t} = K e^{-rt} n(d_-) \sqrt{t}. $$ See Wikipedia.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.