Rough volatility and portfolio optimisation under small transaction costs

Schelling, Denis Matthias (2019) Rough volatility and portfolio optimisation under small transaction costs. PhD thesis, London School of Economics and Political Science.

Text - Submitted Version Download (1MB)

Abstract

The first chapter of the thesis presents the study of the linear-quadratic ergodic control problem of fractional Brownian motion. Ergodic control problems arise naturally in the context of small cost asymptotic expansion of utility maximisation problems with frictions. The optimal solution to the ergodic control problem is derived through the use of an infinite dimensional Markovian representation of fractional Brownian motion as a superposition of Ornstein-Uhlenbeck processes. This solution then allows to compute explicit formulas for the minimised objective value through the variance of the stationary distribution of the Ornstein-Uhlenbeck processes. Building on the first chapter, the second chapter of the thesis presents the main result. This is motivated by the problem an agent faces when trying to minimise her utility loss in the presence of quadratic trading costs in a rough volatility model. Minimising the utility loss amounts to studying a tracking problem of a target that depends on the rough volatility process. This tracking problem is minimised at leading order by an asymptotically optimal strategy that is closely linked to the ergodic control problem of fractional Brownian motion. This asymptotically optimal strategy is explicitly derived. Moreover, the leading order of the small cost expansion is shown to depend only on the roughest part of the considered target. It therefore depends on the Hurst parameter. The third chapter is devoted to a numerical analysis of the utility loss studied in the second chapter. For this, we compare the utility loss in a rough volatility model to a semimartingale stochastic volatility model. The parameter values for both models are fitted to match frictionless utility for realistic values. By applying the result obtained in the second chapter of the thesis, the difference between leading order of utility loss can be explicitly compared.

Item Type:	Thesis (PhD)
Additional Information:	© 2019 Denis Matthias Schelling
Library of Congress subject classification:	H Social Sciences > HG Finance Q Science > QA Mathematics
Sets:	Departments > Mathematics
Supervisor:	Czichowsky, Christoph
URI:	http://etheses.lse.ac.uk/id/eprint/4037

Actions (login required)

Record administration - authorised staff only