Risk Estimation on High Frequency Financial Data, 2015 Empirical Analysis of the DAX 30 BestMasters Series

By studying the ability of the Normal Tempered Stable (NTS) model to fit the statistical features of intraday data at a 5 min sampling frequency, Florian Jacobs extends the research on high frequency data as well as the appliance of tempered stable models. He examines the DAX30 returns using ARMA-GARCH NTS, ARMA-GARCH MNTS (Multivariate Normal Tempered Stable) and ARMA-FIGARCH (Fractionally Integrated GARCH) NTS. The models will be benchmarked through their goodness of fit and their VaR and AVaR, as well as in an historical Backtesting.

Multivariate Standard Normal Tempered Stable Distribution.- FIGARCH.- High Frequency Data and Risk Management.

Florian Jacob obtained his Master’s Degree in Business Engineering from the Karlsruhe Institute of Technology focusing on the application of tempered stable distributions on financial data and financial engineering.