Servicenavigation

Research / Areas of Work

Period search in unevenly sampled time series occuring in astroparticle physics (light curves) using robust methods and also in presence of red noise:

• R package to robustly calculate periodograms, detect periods and simulate light curves possibly containing red noise: RobPer (available on CRAN)
• Documentation on RobPer: Thieler, Fried and Rathjens (2013)
• More details on the methods and the implementation in RobPer: Thieler (2014) (Thesis, in German)
• Earlier research: Thieler, Backes, Fried and Rhode (2013)

 Image filter based on robust regression to smooth images and detect edges:

• A comparison of robust smoothers: Thieler (2009) (Diploma thesis, in German)
• Approaches to detect edges based on robust time series analysis: Raabe, Thieler, Weihs, Rautert and Biermann (2012) and Fried, Raabe and Thieler (2012)

Publications

• Thieler, A.M. (2014): Robuste Verfahren zur Periodendetektion in ungleichmäßig beobachteten Lichtkurven, Thesis, TU Dortmund University. Link
• Thieler, A. M., Fried, R., Rathjens, J. (2013): RobPer: An R package to calculate periodograms for light curves based on robust regression, Technical Report from the Collaborative Research Center SFB 876, TU Dortmund University, Nr. 2 (submitted). Link
• Thieler, A.M., Backes, M., Fried, R., Rhode, W. (2013), "Periodicity detection in irregularly sampled light curves by robust regression and outlier detection", Statistical Analysis and Data Mining 6 (1), 73–89. Link
• Raabe, N., Thieler, A.M., Weihs, C., Fried, R., Rautert, C., Biermann, D. (2012), "Modeling Material Heterogeneity by Gaussian Random Fields for the Simulation of Inhomogeneous Mineral Subsoil Machining", Proceedings of SIMUL 2012, 97-102. Link
• Fried, R., Raabe, N., Thieler, A. (2012), "On the robust analysis of periodic nonstationary time series", Proceedings in Computational Statistics COMPSTAT 2012, 245-257.
• Thieler, A. M. (2009): Filtermethoden auf Basis robuster Regression zur Glättung von Graustufenbildern, Diploma Thesis, TU Dortmund University.