Abstract:We introduce a Markov-functional approach to construct local volatility models that are calibrated to a discrete set of marginal distributions. The method is inspired by and extends the volatility interpolation of Bass (1983) and Conze and Henry-Labordère (2022). The method is illustrated with efficient numerical algorithms in the cases where the constructed local volatility functions are: (1) time-homogeneous between or (2) continuous across, the successive maturities. The step-wise time-homogeneous construction produces a parsimonious representation of the local volatility term structure.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is how to construct a local volatility model so that it can be calibrated to a set of discrete marginal distributions. Specifically, the author proposes a Markov - functional - based method to construct a local volatility model, which can interpolate given a series of marginal distributions. The main contribution of the paper lies in extending the work of Bass (1983) and Conze and Henry - Labordère (2022), providing a new and more flexible way to deal with the volatility interpolation problem.
### Main Problem Description
In the pricing of financial derivatives, the interpolation and extrapolation of the implied volatility surface is a non - trivial problem. There are usually two main methods to solve this problem:
1. **Best - fit of parametric models**: such as the Heston model or the SABR model.
2. **Non - parametric methods**: including the two - step method (first fitting the implied volatility skew in the strike price dimension, and then interpolating in the maturity dimension) and the global non - parametric representation (directly fitting all observed option prices).
However, these methods face challenges in practical applications because only a limited number of option price data points can be observed in the market. Therefore, how to construct a reasonable local volatility model from these discrete data points to ensure that it is consistent with market observation data and can smoothly transition between different maturities has become a key problem.
### Specific Solutions in the Paper
The paper proposes a new method based on the Markov - functional framework to construct a local volatility model by introducing a local drift term. The specific steps are as follows:
1. **Define the Markov - functional model**:
- Set a target variable \( S_t \) and a driving variable \( X_t \), where \( S_t = f(t, X_t) \), and \( X_t \) is driven by Brownian motion and guided by a local drift function \( \mu(t, x) \).
- Use the Fokker - Planck equation to describe the evolution of the forward density \( p(t, x) \) of \( X_t \).
2. **Solve the flow function and the drift function**:
- Derive the partial differential equation satisfied by the flow function \( f(t, x) \) through the Itô formula.
- Solve \( f(t, x) \) and \( \mu(t, x) \) using boundary conditions or terminal conditions to satisfy the given marginal distributions.
3. **Quantile matching**:
- At each maturity \( T_i \), use the quantile - matching method to determine the flow function \( f(T_i, x) \), that is, \( f(T_i, x) = F^{-1}_{\nu_i} \circ F_{X_{T_i}}(x) \).
4. **Continuity treatment**:
- Propose two methods, the step - by - step time - homogeneous diffusion process and the continuous flow, to ensure the continuity of local volatility between maturities.
### Summary
The core problem of the paper is to construct a local volatility model that can be accurately calibrated to discrete marginal distributions by introducing a local drift term and using the Markov - functional framework. This method not only solves the problems of discontinuity and unreasonable shape in traditional methods, but also provides more flexible and efficient numerical algorithms to achieve this goal.