ResearchI am interested in statistical methods for the analysis of data that defies traditional assumptions (such as Gaussianity or stationarity). In my work I combine classical tools such as quantile regression, copulas and spectral analysis of time series to shed new light on temporal dependence. The procedures I propose are legitimised by rigorous mathematical theory and implemented as statistical software available as R packages.
My papers can be found on Google Scholar. My software can be found on GitHub and on depsy.org.
Conferences I will be attending: Statistical scalability, 4th ISNPS conference 2018.
Work in progress
- Sequential detection of structural changes in irregularly observed data. (with P. Fryzlewicz).
- Predictive, finite-sample model choice for time series under stationarity and non-stationarity. submitted. (with P. Preuß and P. Fryzlewicz). [paper, supplement, arXiv, CRAN, R code*]
- Quantile cross-spectral measures of dependence between economic variables. submitted. (with J. Baruník) [arXiv, SSRN, R code*].
- On Wigner-Ville spectra and the unicity of time-varying quantile-based spectral densities. Journal of Time Series Analysis, to appear. (with S. Birr, H. Dette, M. Hallin, and S. Volgushev) [PDF, arXiv].
- Quantile spectral analysis for locally stationary time series. Journal of the Royal Statistical Society: Series B, 79(5), 1619-1643. (2017, with S. Birr, H. Dette, M. Hallin, and S. Volgushev) [PDF, arXiv].
- Quantile spectral processes: asymptotic analysis and inference. Bernoulli, 22(3), 1770-1807. (2016, with H. Dette, M. Hallin, and S. Volgushev) [PDF, arXiv].
- Quantile-based spectral analysis in an object-oriented framework and a reference implementation in R: the quantspec Package. Journal of Statistical Software, 70(3), 1-27. (2016) [PDF, arXiv]
- Of copulas, quantiles, ranks and spectra: an L1-approach to spectral analysis. Bernoulli, 21(2), 781-831. (2015, with H. Dette, M. Hallin, and S. Volgushev). [PDF, arXiv]
- forecastSNSTS: Forecasting for Stationary and Non-Stationary Time Series. R package version 1.2-0. [CRAN, GitHub]
- quantspec: Quantile-Based Spectral Analysis Functions. R package version 1.2-1. [CRAN, GitHub]
- Quantile-based spectral analysis: asymptotic theory and computation. [PDF]
Previous employmentI have been employed at the Department of Statistics of London School of Economics, Westfälische Wilhelms-Universität Münster and Ruhr-Universität Bochum before I came to work at LSE.
Humboldt-Universität zu Berlin
School of Business and Economics
Unter den Linden 6
Telephone: (49-30) 2093-99412