Multiscaling in the Rough Bergomi Model: A Tale of Tails

An arXiv quantitative finance paper investigates the source of multiscaling behavior in the rough Bergomi (rBergomi) model. It proposes a two-stage statistical testing procedure that uses shuffled surrogates preserving the return distribution while destroying temporal correlations, combined with synthetic processes such as multifractal random walks and fractional Lévy stable motions, to verify robustness. The study finds that multiscaling in the model mainly arises from the heavy-tailed return distribution rather than memory effects induced by the roughness of the volatility path.