Volatility density estimation by multiplicative deconvolution

TL;DR

This paper introduces a non-parametric method for estimating the density of an unobserved volatility process in stochastic models using multiplicative deconvolution and Mellin transforms, with proven consistency and practical implementation.

Contribution

It develops a novel estimator based on Mellin transform inversion and spectral regularisation for volatility density estimation from discrete data, including a data-driven parameter choice.

Findings

Estimator is consistent under regularity conditions.

Provides upper bounds for mean integrated squared risk.

Simulation study demonstrates estimator's practical performance.

Abstract

We study the non-parametric estimation of an unknown stationary density fV of an unobserved strictly stationary volatility process $(\bm V_t)_{t\geq 0}$ on $\IRp^2 := (0,\infty)^2$ based on discrete-time observations in a stochastic volatility model. We identify the underlying multiplicative measurement error model and build an estimator based on the estimation of the Mellin transform of the scaled, integrated volatility process and a spectral cut-off regularisation of the inverse of the Mellin transform. We prove that the proposed estimator leads to a consistent estimation strategy. A fully data-driven choice of $\bm k \in \IRp^2$ is proposed and upper bounds for the mean integrated squared risk are provided. Throughout our study, regularity properties of the volatility process are necessary for the analsysis of the estimator. These assumptions are fulfilled by several examples of volatility processes which are listed and used in a simulation study to illustrate a reasonable behaviour of the proposed estimator.