Joint Calibration of the Five-Parameter SVI Volatility Smile

Summary

The document asks whether SVI volatility-surface calibration should follow a two-stage procedure, first fitting the parameters a, b, and rho, then fitting m and sigma, or whether the order can be reversed. The response recommends optimizing all five parameters together instead of splitting them into successive fits. Its reasoning is that a joint fit can better minimize the calibration objective, and that the optimization was reported to run quickly in an implementation handling many volatility smiles.

The answer also distinguishes joint calibration from approaches that fix selected parameters in advance. It uses SABR as an analogy: traders may set a backbone parameter, leaving fewer parameters for optimization. This is practical guidance rather than a derivation or a systematic comparison of calibration procedures. The document gives no objective function, constraints, initialization method, data set, or error metrics, so it does not establish that simultaneous optimization is always more stable or produces a better usable surface in every setting.

Key ideas

• The response recommends fitting all five SVI parameters jointly rather than in two sequential groups.
• Joint fitting can directly minimize the chosen calibration objective across the full parameter set.
• Some parameters may instead be fixed using market or modeling assumptions before calibration.
• The reported speed comes from one implementation and does not establish performance for every setup.

Tags

# SVI Zeliade Vol Surface Calibration


# SVI Zeliade Vol Surface Calibration












Have a question about SVI Zeliade Implementation (pdf, overview). The paper suggested to do 2 rounds of optimization, first for $\{a,b,\rho\}$ and 2nd for $\{m,\sigma\}$.

Does anyone know if I can invert the order of optimization. First, optimize $\{m,\sigma\}$, and then optimize $\{a,b,\rho\}$.

Benedict

## Answer by Hui (score -2)

https://quant.stackexchange.com/a/39774

I didn't find the methodologies you mentioned. My thought is you should do the five parameters optimization at the same time. Don't be scared by 5 parameters. The minimization process is actually pretty fast. We implemented in Java and to optimize SVI parameters for thousands of vol smiles take only 1 minute. Minimizing 5 parameters will definitely provide you with best fit than do them separately.

I have seen some methods that fix values on some parameters but never heard of minimizing a function separately (I could definitely be wrong). In SABR model, the backbone could be set by traders, so only three parameters could be involved in minimization although it has 4 parameters totally. It could be similar case in SVI as well to fix some parameters but not for two times minimizations

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.