Quick bit of background. I'm adding a little options market making into my normal crypto derivatives trading, and I'm currently writing the software to help me do that effectively. One thing I'm missing is being able to effectively model the volatility smile in isolation. I can fit my curve to the market, but then I won't know if a particular contract is over or under the theoretical value. Any suggestions on how to do it? Or suggestions on where to look for more information? I'm not opposed to buying something if it needs to be a commercial model.
If you've fitted some sort of spline to the market prices wouldn't you be left with residuals where the market prices don't match the fit exactly?
Yes Thanks for the reply! But I still need the market data in the first place to fit the spline. I guess there's nothing but market forces and experience that dictates the exact shape of the curve. I don't have the experience yet to visualise the curve in abstraction, so maybe I'll stick to fitting to the market and picking off outliers!
Yes that's all there is really. A market-maker would just "try" a given vol-curve and then adjust the shape of the curve they bid/offer around given the fills they get.