Stochastic Volatility for Derivatives Pricing and Hedging

# Why does a trader need stochastic volatility models?


# Why does a trader need stochastic volatility models?












I am trying to understand the practical motivation for stochastic volatility models from the point of view of a derivatives trader.

For example, a digital put can be (in theory) statically hedged using a tight put spread: \begin{equation} \frac{1}{|K_2-K_1|}\mathrm{PS}(K_1,K_2) \to \mathrm{DP}\left(\frac{K_1+K_2}{2}\right), \end{equation} but markets only quote a discrete set of strikes, so static replication is imperfect. Thus, the trader needs to add a premium for the risks he is not hedging with this approximate strategy. How are these premiums computed?

My current understanding of the workflow is the following:

- A client sends a request for quote through Bloomberg:

> client: Price for a 1W @5525 digital put on SX5E, notional 1M.

- The trader reviews their book: checks current Greeks, decides which risks they want to take on and which they prefer to avoid.

- They run the pricing engine (developed by the quant team) for the digital put and decide how they would hedge the trade.

- They estimate the expected cost of that hedge and the additional margins required for their profit.

With all that, the trader builds a quote and replies to the client:

> trader: 1W @5525 digital put on SX5E at 3.60% notional.

In this sense, a pricing model is truly useful to a trader when it provides a clear decomposition of the risk. The better it identifies what is being hedged and what the real cost of that hedge is, the more solid the quoted premium will be and the easier it becomes for the trader to keep only the risks that fit their book.

Given the structure of the market (volatility surface, skew dynamics, vol-of-vol, spot–vol correlation, etc.), traders must distinguish the different components of volatility risk, so stochastic volatility models seem essential for determining these premiums.

My question:

Is this the correct motivation for stochastic volatility models from a trading/hedging perspective? And how are these risk premiums actually computed in practice?

## Answer by QuantCalc.net (score 1)

https://quant.stackexchange.com/a/85276

Short answer is volatility clustering and Greeks stability. Stochastic volatility models are used because real markets exhibit volatility clustering—periods of persistently high or low volatility—and because they provide much more stable and realistic Greeks for hedging. Unlike constant-volatility models, SV models generate the observed volatility smile/skew, capture fat tails and the leverage effect, and produce smoother Delta, Gamma, and Vega that align with how the implied-vol surface actually moves. This improves hedge effectiveness, reduces P/L noise, and better reflects how volatility evolves in clusters, making stochastic volatility a far more accurate and practical framework for pricing and risk management.