Podcast: Mathieu Rosenbaum on the rough Heston model

Combination of rough volatility and the classical Heston model gives promising results

Nazneen Sherif and Mauro Cesa

Photo: Monika Ghose

This Quantcast episode is dedicated to rough volatility models, which originated from attempts to model volatility without assuming its surface is smooth, as implied by, for example, the Black-Scholes model.

Our guest, Mathieu Rosenbaum, head of analytics and models for regulation at the department of applied mathematics of the École Polytechnique, is one of the authors who introduced the concept of rough volatility to finance in 2014, together with Jim Gatheral and Thibault Jaisson (Gatheral et al, 2014), and has contributed to develop models in this area since.

According to Rosenbaum, these models are set to become more popular because of their ability to describe markets’ volatility accurately. “It is very hard to reject that volatility is rough on historical data and it is very hard to reject that volatility is rough on implied volatility data,” says Rosenbaum.

In addition, he points out their specification is very parsimonious. There are only three parameters: volatility level, volatility of volatility and spot/vol correlation.

His latest work, Roughening Heston, is co-authored with Omar El Euch and Jim Gatheral. It extends the original rough volatility model combining it with the classical Heston model.

Rosenbaum explains the issues related to applicability and implementation of the rough Heston model. Also, he offers his answers to some of the criticisms the model has received, such as the limited applicability of a non-Markovian model and the precision of the fitting of a model with only three parameters.

INDEX

00:00 Intro

01:15 What is rough volatility?

02:25 Development of the model

04:15 Advantages of rough volatility

06:32 Rough Heston paper

09:02 Implementation

11:25 Rough volatility with other volatility models

12:50 What products can rough volatility be applied to

13:35 Limited applicability of non-Markovian models?

16:25 Poor man’s rough Heston model

18:24 Fitting precision

21:20 Are banks using it?

 

References

Jim Gatheral, Thibault Jaisson, Mathieu Rosenbaum (2014)
Volatility is rough
https://arxiv.org/abs/1410.3394

 

To hear the full interview, listen in the player above, or download. Future podcasts in our Quantcast series will be uploaded to Risk.net. You can also visit the main page here to access all tracks, or go to the iTunes store or Google Podcasts to listen and subscribe.