Emil S. Jørgensen

PHD research fellow, University of Copenhagen

Emil is a PhD research fellow at the University of Copenhagen in Denmark. He will be visiting the Stevanovich Center for a duration of 6 months, starting March 2016.

His primary research interest is high-frequency statistics, in particular estimation problems for diffusion-type processes with and without jumps.

For his first paper, Emil is developing asymptotic methods for inference and testing of a particular time-changed diffusion process that will be used to model the dynamics of a world stock index. The model is a stochastic volatility model, driven only by the continuous uncertainty of the underlying market activity. High-frequency limit theorems are derived for the estimation error of the unknown parameters. Emphasis is on model validation issues, which will include the construction and application of a nonparametric test for whether the latent volatility process is driven by the same Brownian motion as the index, or a linear combination with an independent factor.

His second project is entitled "Inference for integrated diffusions observed at high frequency". The aim of the project is two-fold; firstly, to derive limit theorems for estimators obtained from prediction-based estimating functions (PBEFs) within a high-frequency asymptotic framework and, secondly, to illustrate the results by constructing simulation-based PBEFs for the square-root (CIR) model and the 3/2 diffusion model. The latter is a popular model in finance for which explicit Godambe-Heyde optimal PBEFs are not readily available, which motivates the simulation-based extension of the general theory. The results are of particular interest for stochastic volatility modeling in finance, where realized volatility or variations thereof may be used to construct a trajectory of the integrated volatility process.

In a third project, Emil studies parameter estimation for finite activity jump diffusions observed in a novel high-frequency asymptotic framework, where the jump intensity increases with the number of observations. The working title of the paper is "Finite horizon λ-asymptotics”. New identifiability notions are defined to match this particular asymptotic scenario.

 

Before joining University of Copenhagen in 2014, Emil obtained an honorary Master’s degree in Mathematical Modeling and Computation from the Technical University of Denmark.