How do you apply the theory of rough paths in the context of quantitative finance, and what are its advantages over classical stochastic calculus?

The theory of rough paths provides a mathematical framework to extend classical stochastic calculus to a broader class of "rough" signals or irregular paths. In quantitative finance, this theory has been utilized to model and analyze financial time series and various aspects of financial markets, such as volatility, that exhibit rough or irregular behavior.

One of the main advantages of using rough path theory instead of classical stochastic calculus is that it can deal with processes that exhibit non-regular behavior such as fractional Brownian motion, which has long-range memory, and may be a more suitable representation of financial markets compared to regular Brownian motion.

Let us consider an example to illustrate the application of rough path theory in quantitative finance. In the rough volatility framework, the asset price dynamics are driven by a fractional Brownian motion, which allows for a more realistic modeling of implied volatility surfaces. The asset price dynamics are described by the following equation:

dS_t = μ_tS_tdt + σ_tS_tdW_t^H,

where S_t is the asset price, μ_t is the drift, σ_t is the volatility, dW_t^H represents the fractional Brownian motion with Hurst parameter H ∈ (0, 1) and dt denotes time. To study such a system, we consider the signature of the volatility process, X_t = (σ_t, W_t^H), which forms a rough path.

To further demonstrate the application of rough path theory in quantitative finance, let’s consider the two widely known stochastic volatility models – the Heston model and the rough Heston model.

**Heston Model (Classical Stochastic Calculus):**

The Heston model represents the price process via the following stochastic differential equations (SDEs):

$$\begin{aligned} dS_t &= S_t\,(\mu dt + \sqrt{v_t}\,dW^{(1)}_t)\\ dv_t &= \kappa(\theta - v_t)\,dt + \eta\sqrt{v_t}\,dW^{(2)}_t\\ \end{aligned}$$

Here, S_t denotes the asset price, v_t is the instantaneous variance, μ is the drift, κ is the mean-reversion rate, θ is the long-term mean variance, η represents the volatility of volatility, and W_t⁽¹⁾ and W_t⁽²⁾ are correlated Brownian motions with correlation ρ.

**Rough Heston Model (Rough Path Theory):**

In the rough Heston model, we introduce roughness in the volatility dynamics by using fractional Brownian motion W_t^H. The model is given by:

$$\begin{aligned} dS_t &= S_t\,(\mu dt + \sqrt{v_t}\,dW^{(1)}_t)\\ v_t &= V_0 + \int_0^t g(t-s)\,d(\kappa, \eta W^{(2)}_{s})^H, \end{aligned}$$

where g(t − s) is an appropriate kernel function, and (κ, ηW_s⁽²⁾)^H denotes a fractional Brownian motion with Hurst parameter H ∈ (0, 1). The integral with respect to the fractional Brownian motion can be approached using rough paths and the so-called "Volterra Rough Integral", which is a special type of rough integral.

In summary, the theory of rough paths offers several advantages over classical stochastic calculus in quantitative finance, such as handling non-regular behavior of financial time series, better modeling of financial markets, and the ability to analyze rough volatility models. By incorporating long-range memory and other non-Markovian characteristics of financial markets, rough path theory can provide more accurate and insightful quantitative analysis of financial phenomena.