Alwin Stegeman

SCIENCE

career

My work on scientific research mostly took place at the University of Groningen, The Netherlands, from which I graduated in Econometrics (1998) and Mathematical Statistics (1998), and obtained my PhD on limit theorems for stochastic processes (2002). After one year of Operations Research (2003) I moved to Psychometrics where I received VENI (2005) and VIDI (2008) research grants from the Netherlands Organisation for Scientific Research (NWO) for my research on tensor decompositions. In 2015 I left the department of Psychometrics as tenured associate professor. In 2016 I started working as visiting professor at KU Leuven, campus Kortrijk, Belgium, at the department of Engineering. There I joined a project to solve polynomial systems by using tensor decompositions. I received an EU Marie Curie fellowship (2017) to continue this research, but I decided to explore life outside academia instead.

​Below you can find my publications and technical reports, including accompanying Matlab codes, and slides of some of my presentations. If you believe in citation scores, then this may be of interest ;-)

publications
  1. A. Stegeman & L. De Lathauwer (2019). Rayleigh quotient methods for estimating common roots of noisy univariate polynomials. Computational Methods in Applied Mathematics, Special issue on Tensor numerical methods: theoretical analysis and modern applications, 19, 147-163.
  2. A. Stegeman (2018). Simultaneous component analysis by means of Tucker3. Psychometrika , 83, 21-47. [pdf] [matlab codes]
  3. I. Domanov, A. Stegeman, L. De Lathauwer (2017). On the largest multilinear singular values of higher-order tensors. SIAM Journal on Matrix Analysis and Applications, 38, 1434-1453. [pdf]
  4. A. Stegeman & S. Friedland (2017). On best rank-2 and rank-(2,2,2) approximations of order-3 tensors. Linear and Multilinear Algebra, 65, 1289-1310. [pdf]
  5. B.M. Armenta, K. Stroebe, S. Scheibe, N.W. Van Yperen, A. Stegeman, T. Postmes (2017). Permeability of group boundaries: development of the concept and a scale. Personality and Social Psychology Bulletin, 43, 418-433. [pdf]
  6. A. Stegeman (2016). A new method for simultaneous estimation of the factor model parameters, factor scores, and unique parts. Computational Statistics & Data Analysis, 99, 189-203. [pdf] [matlab codes]
  7. I.A.M. Smits, M.E. Timmerman & A. Stegeman (2016). Modelling non-normal data: the relationship between the skew-normal factor model and the quadratic factor model. British Journal of Mathematical and Statistical Psychology, 69, 105-121. [pdf]
  8. A. Stegeman & T.T.T. Lam (2016). Multi-set factor analysis by means of Parafac2. British Journal of Mathematical and Statistical Psychology, 69, 1-19. [pdf]
  9. R. Monden, A. Stegeman, H.J. Conradi, P. de Jonge & K.J. Wardenaar (2016). Predicting long-term depression outcome using a three-mode principal component model for depression heterogeneity. Journal of Affective Disorders, 189, 1-9. [pdf]
  10. R. Monden, K.J. Wardenaar, A. Stegeman, H.J. Conradi & P. de Jonge (2015). Simultaneous decomposition of depression heterogeneity on the person-, symptom- and time-level: the use of three-mode principal component analysis. PloS ONE, 10(7): e0132765. [pdf]
  11. K. Stroebe, T. Postmes, S. Täuber, A. Stegeman & M.-S. John (2015). Belief in a just what? Demystifying just world beliefs by distinguishing sources of justice. PloS ONE, 10(3): e0120145. [pdf]
  12. A. Stegeman (2014). Finding the limit of diverging components in three-way Candecomp/Parafac - a demonstration of its practical merits. Computational Statistics & Data Analysis, 75, 203-216. [pdf] [matlab codes]
  13. A. Stegeman & T.T.T. Lam (2014). Three-mode factor analysis by means of Candecomp/Parafac. Psychometrika, 79, 426-443. [pdf]
  14. A. Stegeman (2013). A three-way Jordan canonical form as limit of low-rank tensor approximations. SIAM Journal on Matrix Analysis and Applications, 34, 624-650. [pdf] [matlab codes]
  15. A. Stegeman & T.T.T. Lam (2012). Improved uniqueness conditions for canonical tensor decompositions with linearly dependent loadings. SIAM Journal on Matrix Analysis and Applications, 33, 1250-1271. [pdf]
  16. A.L.F. de Almeida, X. Luciani, A. Stegeman & P. Comon (2012). CONFAC decomposition approach to blind identification of underdetermined mixtures based on generating function derivatives. IEEE Transactions on Signal Processing, 60, 5698-5713. [pdf]
  17. A. Stegeman (2012). Candecomp/Parafac: from diverging components to a decomposition in block terms. SIAM Journal on Matrix Analysis and Applications, 33, 291-316. [pdf] [matlab codes]
  18. X. Guo, S. Miron, D. Brie, & A. Stegeman (2012). Uni-mode and partial uniqueness conditions for Candecomp/Parafac of three-way arrays with linearly dependent loadings. SIAM Journal on Matrix Analysis and Applications, 33, 111-129. [pdf]
  19. A. Stegeman (2011). On uniqueness of the canonical tensor decomposition with some form of symmetry. SIAM Journal on Matrix Analysis and Applications, 32, 561-583. [pdf]
  20. A. Stegeman & P. Comon (2010). Subtracting a best rank-1 approximation may increase tensor rank. Linear Algebra and its Applications, 433, 1276-1300. [pdf]
  21. A. Stegeman (2010). On uniqueness of the n-th order tensor decomposition into rank-1 terms with linear independence in one mode. SIAM Journal on Matrix Analysis and Applications, 31, 2498-2516. [pdf]
  22. A. Stegeman & A.L.F. de Almeida (2009). Uniqueness conditions for constrained three-way factor decompositions with linearly dependent loadings. SIAM Journal on Matrix Analysis and Applications, 31, 1469-1490. [pdf]
  23. A. Stegeman (2009). On uniqueness conditions for Candecomp/Parafac and Indscal with full column rank in one mode. Linear Algebra and its Applications, 431, 211-227. [pdf]
  24. A. Stegeman (2009). Using the Simultaneous Generalized Schur Decomposition as a Candecomp/Parafac algorithm for ill-conditioned data. Journal of Chemometrics, Special issue in memory of Richard Harshman, 23, 385-392. [pdf]
  25. A. Stegeman & L. De Lathauwer (2009). A method to avoid diverging components in the Candecomp/Parafac model for generic I×J×2 arrays. SIAM Journal on Matrix Analysis and Applications, 30, 1614-1638. [pdf]
  26. J.M.F. ten Berge, A. Stegeman & M. Bennani Dosse (2009). The Carroll & Chang conjecture of equal Indscal components when Candecomp/Parafac gives perfect fit. Linear Algebra and its Applications, 430, 818-829. [pdf]
  27. W.P. Krijnen, T.K. Dijkstra & A. Stegeman (2008). On the non-existence of optimal solutions and the occurrence of "degeneracy" in the Candecomp/Parafac model. Psychometrika, 73, 431-439. [pdf]
  28. A. Stegeman (2008). Low-rank approximation of generic p×q×2 arrays and diverging components in the Candecomp/Parafac model. SIAM Journal on Matrix Analysis and Applications, Special issue on Tensor Decompositions and Applications, 30, 988-1007. [pdf]
  29. A. Stegeman (2007). Degeneracy in Candecomp/Parafac and Indscal explained for several three-sliced arrays with a two-valued typical rank. Psychometrika, 72, 601-619. [pdf]
  30. A. Stegeman & N.D. Sidiropoulos (2007). On Kruskal's uniqueness condition for the Candecomp/Parafac decomposition. Linear Algebra and its Applications, 420, 540-552. [pdf]
  31. J.M.F. ten Berge & A. Stegeman (2006). Symmetry transformations for square sliced three-way arrays, with applications to their typical rank. Linear Algebra and its Applications, 418, 215-224. [pdf]
  32. A. Stegeman (2006). Degeneracy in Candecomp/Parafac explained for p×p×2 arrays of rank p+1 or higher. Psychometrika, 71, 483-501. [pdf]
  33. A. Stegeman, J.M.F. ten Berge & L. De Lathauwer (2006). Sufficient conditions for uniqueness in Candecomp/Parafac and Indscal with random component matrices. Psychometrika, 71, 219-229. [pdf]
  34. A. Stegeman & J.M.F. ten Berge (2006). Kruskal's condition for uniqueness in Candecomp/Parafac when ranks and k-ranks coincide. Computational Statistics & Data Analysis, 2nd Special issue on Matrix Computations and Statistics, 50, 210-220. [pdf]
  35. W.K. Klein Haneveld & A.W. Stegeman (2005). Crop succession requirements in agricultural production planning. European Journal of Operational Research, 166, 406-429. [pdf]
  36. A. Stegeman (2002). Extremal behavior of heavy-tailed ON-periods in a superposition of ON/OFF processes. Advances in Applied Probability, 34, 179-204. [pdf]
  37. T. Mikosch, S. Resnick, H Rootzén & A. Stegeman (2002). Is network traffic approximated by stable Lévy motion or fractional Brownian motion? Annals of Applied Probability, 12, 23-68. [pdf]
  38. A. Stegeman (2000). Heavy tails versus long-range dependence in self-similar network traffic. Invited publication. Statistica Neerlandica, 54, 293-314. [pdf]
  39. A. Stegeman (2002). The Nature of the Beast, Analyzing and Modeling Computer Network Traffic. PhD Thesis, Department of Mathematics, University of Groningen, The Netherlands. [pdf]
reports
  • A. Stegeman (2015). On the (non)existence of best low-rank approximations of generic I×J×2 arrays. [arXiv]
  • A. Stegeman & L. De Lathauwer (2011). Are diverging CP components always nearly proportional? [arXiv]
  • A. Stegeman (2010). The Generalized Schur Decomposition and the rank-R set of real I×J×2 arrays. [arXiv]
  • A. Stegeman & A. Mooijaart (2008). Independent Component Analysis with Errors by Least Squares Covariance Fitting. (former title: Independent Factor Analysis by Least Squares). [pdf]
  • A. Stegeman (2007). Comparing Independent Component Analysis and the Parafac model for artificial multi-subject fMRI data. [pdf]
  • A. Stegeman (2005). Degeneracy in Candecomp/Parafac explained for 5×3×3 arrays of rank 6 or higher. [pdf]
  • A. Stegeman & J.M.F. ten Berge (2004). Real-valued 4×3×3 arrays have rank 5 with positive probability. [pdf]
  • A. Stegeman (2001). Non-stationarity versus long-range dependence in computer network traffic. [postscript 2002]
slides
  • A. Stegeman (2016). A three-way Jordan canonical form as limit of low-rank tensor approximations. Workshop on Tensor Decompositions and Applications (TDA 2016), January 18-22, Leuven, Belgium. [pdf]
  • A. Stegeman (2015). Decomposing a three-way dataset of TV-ratings when this is impossible. Invited speaker at Maastricht University, Department of Knowledge Engineering (DKE), October 28, Maastricht. [pdf]
  • A. Stegeman (2015). Direct-Fitting Common Factor Analysis. VOC meeting, May 29, Nijmegen. [pdf]
  • A. Stegeman (2014). Factor indeterminacy and its implications for Item Response Theory (new version). 5th RCEC Workshop on IRT and Educational Measurement, November 19-21, Enschede, The Netherlands. [pdf]
  • A. Stegeman (2014). Nonexistence of best low-rank approximations for real-valued three-way arrays. 21st International Symposium on Mathematical Theory of Networks and Systems (MTNS 2014), July 7-11, Groningen, The Netherlands. [pdf] [extended abstract]
  • A. Stegeman (2013). 3-way Candecomp/Parafac and the diverging components problem. Conference of the International Federation of Classification Societies (IFCS 2013), July 14-17, Tilburg, The Netherlands. [pdf]
  • A. Stegeman (2013). A solution for diverging components in 3-way Candecomp/Parafac. Research Meeting Psychometrics & Statistics, University of Groningen, March 13. [pdf]
  • A. Stegeman (2010). Candecomp/Parafac - From diverging components to a decomposition in block terms. Should have been a talk at TDA 2010, September 13-17, Bari, Italy. [pdf]
  • A. Stegeman (2006). The Parafac model for multi-way data analysis. Invited tutorial at the 1st International Summer School in Biomedical Engineering, August 7-11, Erfurt, Germany. [pdf]

Contact

stegeman.alwin@gmail.com

LinkedIn

GitHub