LinkedIn Twitter
Research and Publications

Last News

  • The December 2017 issue of Statistical Modelling, an Academic journal for which I'm acting as Associate Editor, is on line. This issue contains papers on count models, integer valued Garch models, Dirichlet processes, additive hazard models, etc.
    Impact factor: 1.074, Ranked 57 out of 124 in Statistics & Probability.
    5-Year impact factor: 1.444, Ranked 52 out of 124 in Statistics & Probability (Source: 2016 Journal Citation Reports)
    Potential authors are encouraged to read the aims and scope of this journal and submit their manuscript.
  • IMT Atlantique-Télécom Bretagne
    Supervision de mémoires de 3ème année sur le machine learning (un groupe), le deep-learning (les deux groupes), et la prédiction "model free" (un groupe) : je suis disponible le mardi, toute la journée, et le jeudi après 16h pour les visioconférences.
  • Mon cours de 3ème année à l'IMT Atlantique-Télécom Bretagne sur les processus non stationnaires est en cours de refonte pour 2017-2018 à la suite d'une série d'articles à analyser pour les Mathematical Reviews : articles MR3210268 à MR3210277, et MR3551912 pour la section sur les modèles à changement de régime, article MR3375197 pour la section sur les estimateurs spot, article MR3311862 sur la multifractalité, et le livre MR3441999 qui traite entre autres de la prévision directement à partir des données sans référence à un modèle spécifique et de la statistique computationelle est d'un grand intérêt pour le traitement statistique des grands échantillons de données (les « big data »). Le livre tiré du cours prend enfin forme. Liste des leçons  pour 2016-2017:
    • Leçon 1 : Modèles de volatilité. Modèle GARCH : définition, propriétés, structure de dépendance, estimation par QML et prévision. Transparents
    • Leçon 2 : Modèle GARCH exponentiel. Inférence des modèles univariés. Modèles multivariés. Transparents
    • Leçon 3 : Modèles de volatilité fortement dépendants. Modèles à changement de régime. Transparents
    • Leçon 4 : Estimation de la volatilité de processus observés à très haute fréquence. Analyse par ondelettes de la volatilité. Multifractalité. Transparents

    I'm currently updating my 3rd year lectures at IMT Atlantique-Télécom Bretagne on nonstationary processes for the academic year 2017-2018. The sections on change-point, spot volatility estimators, multifractality are modified, and a section on computational statistics and “big data” is added after reviewing a set of papers and a book on these issues for Mathematical Reviews: MR3210268 to MR3210277, MR3375197, MR3311862, MR3442999 and MR3551912. The writing of the book supporting these lectures is in progress.
    List of lectures for 2016-2017:
    • Lecture 1: Volatility models. GARCH model: definition, properties, dependence structure, QML estimation, model-based and model-free prediction. Slides
    • Lecture 2: Exponential GARCH model. Inference of univariate models. Multivariate models. Slides
    • Lecture 3: Long-range dependent volatility models. Change-point models. Slides
    • Lecture 4: Volatility estimation of high frequency time series. Wavelet analysis of volatility processes. Multifractality. Slides

Research Interests

  • Time series,
  • Long-memory and multifractal processes, change-point detection,
  • Statistical analysis of financial data, volatility modeling,
  • Financial markets with interacting agents, financial bubbles,
  • Wavelet signal processing of scaling processes,
  • Machine learning, computational statistics, and big data analysis.

Publications

  1. D. Surgailis, G. Teyssière and M. Vaičiulis. The increment ratio statistic.
    Journal of Multivariate Analysis (2008) vol 99, 510-541. DOI. MR2396977. Pdf file (Supplementary Material Pdf).
  2. G. Teyssière and P. Abry. Wavelet analysis of nonlinear long-range dependent processes. Applications to financial time series.
    In Long-Memory in Economics. G. Teyssière et al. editors, 173-238, Springer (2007). DOI. MR2265060. Pdf file.
  3. M. Lavielle and G. Teyssière. Adaptive detection of multiple change-points in asset price volatility.
    In Long-Memory in Economics. G. Teyssière et al. editors, 129-156, Springer (2007). DOI. MR2265058. Pdf file.
  4. D. Kateb, A. Seghier and G. Teyssière. Prediction, orthogonal polynomials and Toeplitz matrices: a fast and reliable approximation to the Durbin-Levinson algorithm,
    In Long-Memory in Economics. G. Teyssière et al. editors, 239-261, Springer (2007). DOI. MR2265061. Pdf file.
  5. M. Lavielle and G. Teyssière. Detection of multiple change-points in multivariate time series.
    Lithuanian Mathematical Journal (2006) vol 46, 287-306. DOI. MR2285348. Pdf file. M. Lavielle and G. Teyssière. Détection de ruptures multiples dans des séries temporelles multivariées (French version of this paper).
    Lietuvos Matematikos Rinkinys (2006) vol 46, 351-376. Pdf file.
  6. P. Doukhan, G. Teyssière and P. Winant. A LARCH$(\infty)$ vector valued process.
    In Dependence in Probability and Statistics. Lecture Notes in Statistics. P. Bertail, P. Doukhan and Ph. Soulier editors, vol 187, 245-258, Springer (2006). DOI. MR2283258. Pdf file.
  7. A. Kirman and G. Teyssière. Testing for bubbles and change-points.
    Journal of Economic Dynamics and Control (2005) vol 29, 765-799. DOI. MR2129522. Pdf and PostScript files.
  8. L. Horváth, P. Kokoszka and G. Teyssière. Bootstrap misspecification tests for ARCH based on the empirical process of squared residuals.
    Journal of Statistical Computation and Simulation (2004) vol 74, 469-485. DOI. MR2073226. Pdf and PostScript files.
  9. P. Kokoszka, G. Teyssière and A. Zhang. Confidence intervals for the autocorrelations of the squares of GARCH sequences.
    In Computational Science - ICCS 2004. Lecture Notes in Computer Science. M. Bubak et al. editors, vol 3039, 827-834, Springer (2004). DOI. MR2233424. Pdf file. Volume for the Workshop on Computational Methods in Finance and Insurance, Kraków, Poland, June 2004. Slides.
  10. L. Giraitis, P. Kokoszka, R. Leipus and G. Teyssière. On the power of $R/S$-type tests under contiguous and semi long-memory alternatives.
    Acta Applicandae Mathematicae (2003) vol 78, 285-299. DOI. MR2024032. Pdf and PostScript files. (Special Issue for the 8th Vilnius Conference on Probability Theory and Mathematical Statistics), Vilnius, Lithuania.
  11. G. Teyssière. Interaction models for common long-range dependence in asset price volatilities.
    Invited chapter in Processes with Long Range Correlations: Theory and Applications. Lecture Notes in Physics. G. Rangarajan and M. Ding editors, vol 621, 251-269, Springer (2003). DOI. Pdf and PostScript files. Invited lecture to the International Conference on Long-Range Dependent Stochastic Processes and their Applications, Bangalore, India, January 2002.
  12. L. Giraitis, P. Kokoszka, R. Leipus and G. Teyssière. Rescaled variance and related tests for long memory in volatility and levels.
    Journal of Econometrics (2003) vol 112, 265-294. DOI. MR1951145. Pdf file. See also L. Giraitis, P. Kokoszka, R. Leipus and G. Teyssière, Corrigendum to "Rescaled variance and related tests for long memory in volatility and levels",
    Journal of Econometrics (2005) vol 126, 571-572. DOI. MR2155635. Pdf file.
  13. A. Kirman and G. Teyssière. Bubbles and Long Range Dependence in Asset Prices Volatilities.
    In Equilibrium, Markets and Dynamics. C.H. Hommes, R. Ramer and C. Withagen editors, 307-327, Springer (2002).
  14. A. Kirman and G. Teyssière. Microeconomic models for long-memory in the volatility of financial time series.
    SNDE (2002) vol 5, 281-302. DOI.
  15. L. Horváth, P. Kokoszka and G. Teyssière. Empirical process of the squared residuals of an ARCH sequence.
    The Annals of Statistics (2001) vol 29, 445-469. DOI. MR1863965. Pdf and PostScript files.
  16. L. Giraitis, P. Kokoszka, R. Leipus and G. Teyssière. Semiparametric estimation of the intensity of long-memory in conditional heteroskedasticity.
    Statistical Inference for Stochastic Processes (2000) vol 3, 113-128. (Special Issue on Limit Theorems and Long-Range Dependence). DOI. MR1819290. Pdf and PostScript files.
  17. G. Teyssière. Multivariate long-memory ARCH modelling for high frequency foreign exchange rates.
    In Proceedings of the Second High Frequency Data in Finance (HFDF-II) Conference, Olsen & Associates, Zurich, (1998). Pdf and PostScript files.

Books

Statistical Methods for the Detection of Money Laundering.
J-P. Brun and G. Teyssière (Under preparation).


Long-Memory in Economics
G. Teyssière and A. Kirman editors, Springer (2007).
DOI, MR2263582
ISBN (Hardcover): 978-3540226949
ISBN (Paperback): 978-3642061547
ISBN (eBook): 978-3540346258


Dependence in Probability and Statistics, Lecture Notes in Statistics, Vol 200.
G. Lang, D. Surgailis and G. Teyssière editors, Springer (2010).
DOI, MR2741808
ISBN (Paperback): 978-3642141034
ISBN (eBook): 978-3642141041


Preprints, Conference Slides & Occasional Papers

  • G. Teyssière and P. Abry. Wavelet multifractal analysis of high-frequency financial data (2010), 10th Vilnius Conference on Probability Theory and Mathematical Statistics
  • D. Surgailis and G. Teyssière. The increment ratio test for unit root under linear observations (2009).
  • P. Bertrand, G. Teyssière and A. Chamoux. Detection of change-Points in the spectral density. With applications to ECG data. (Occasional paper).
    In Proceedings of the EGC 2009 Conference, Workshop "Fouille de données temporelles et analyse de flux de données" (2009) 3-10. Pdf file.
  • G. Teyssière. Détection de ruptures multiples sur des séries chronologiques univariées et multivariées. Application à des données de prix de l'énergie (2008). (Rapport de recherche pour EDF).
  • G. Teyssière. Long-range dependence and multiple change-points in multivariate time series (2007). Slides
    Invited presentation to the International Conference on Statistical Models for Financial Data II, organized by István Berkes and Lajos Horváth at the Institute of Statistics, Graz University of Technology, Graz, Austria, 23-26 May 2007.
  • L. Giraitis, P.M. Robinson and G. Teyssière. Testing for change-point in cyclical and persistent long-memory processes (2005).
  • G. Teyssière. Bubbles, non-stationarity and double long memory (2004).
    Invited presentation to the International Conference on Statistical Models for Financial Data, organized by István Berkes and Lajos Horváth at the Institute of Statistics, Graz University of Technology, Graz, Austria, May 2004.
  • P. Kokoszka and G. Teyssière. Change-point detection in GARCH models: asymptotic and bootstrap tests, PostScript file. Presented to the Invited Paper Meeting of the 54th Session of the International Statistical Institute, August 2003. Slides. Under revision.
  • G. Teyssière. Nonlinear and semiparametric long-memory ARCH (2001).
    Part of the material of this paper appeared in L. Giraitis, P. Kokoszka, R. Leipus and G. Teyssière On the power of R/S-Type tests under contiguous and semi long-memory alternatives, Acta Applicandae Mathematicae (2003), (Special Issue for the 8th Vilnius Conference on Probability Theory and Mathematical Statistics) vol 78, 285-299. DOI. The remainder of this paper has been inserted in others papers.
  • G. Teyssière. Modelling exchange rates volatility with multivariate long-memory ARCH processes (1997). Pdf and PostScript files (Old Version). Under revision/transformation.
  • G. Teyssière. Double long-memory financial time series (1996), Preprint
    Presented in the 1997 Econometric Society European Meeting, the 1997 Society for Economic Dynamics conference Oxford, the 28th Workshop of the Euro Working Group on Financial Modelling, Vilnius, May 2001, Slides.

    In 1996, I pioneered the class of double long memory processes with the ARFIMA-FIGARCH; this was my first paper in time series analysis. My 1998 paper on multivariate (trivariate) ARFIMA-FIGARCH, published in the proceedings of the High Frequency Data in Finance-II conference organized by Olsen & Associates (see above in the list of publications) is available here in both Pdf and PostScript formats.

Co-Authors