Humboldt-Universität zu Berlin - High Dimensional Nonstationary Time Series

Michael Althof



althofmi [at] hu-berlin [dot] de


+49 30 2093-5630


Office hours:

Dorotheenstr. 1
Upon agreement

Postal address:

IRTG 1792
School of Business and Economics
Humboldt-Universität zu Berlin
Unter den Linden 6
10099 Berlin, Germany


2018 - present

PhD student in Statistics, Humboldt-Universität zu Berlin


Certificate in Quantitative Finance, CQF Institute (Fitch Learning)


ESCP Europe, Paris, Master, Focus on Finance


Vordiplom, Universität zu Köln, Business Management 



Working Papers

Mihoci, A., Althof, M., Chen, C. , and Härdle, W. (2019). FRM Financial Risk Meter, IRTG 1792

Ren R, Althof, M, Härdle WK (2021) Tail Risk Network Effects in the Cryptocurrency Market during the COVID-19 Crisis, Risk Analysis, Wiley Publishers, submitted 20201125

Ben Amor S, Althof M, Härdle WKH (2021) FRM Financial Risk Meter for Emerging Markets, submitted 20210429  "Under review"

Work in Progress

Extension of FRM to the Chinese Region and Macroeconomic Risk Variable Significance

Neural Network Based Approach to Yield Curve Modelling and Asset Allocation

Research Interest

  • Applied Machine Learning (Deep Learning, Unstructured Data Analysis and Modelling)
  • Time Series Forecasting using Machine Learning Tools, Focus on Yield Curves and Crypto Currencies
  • Combine Neural Networks with Stochastic Processes, Focus on Yield Curves and Crypto Currencies
  • Amalgamation of Machine Learning Related Analysis to Robust Portfolio Construction, Focus on Crpto Currency Allocation


  • VL Statistical Tools in Finance and Insurance (WS 19/20)
  • VL Statistics of Financial Markets I - Interest Rate Models (WS 21/22)