LinkedIn Twitter
Teaching and Supervision Material

Online Teaching Material

  • 2019-2020: The list of dissertations topics on machine learning, deep-learning, and "model free" prediction will be available this September.
  • 2010-: Institut Mines-Télécom Atlantique, Brest, lectures (in French) on nonstationary processes.
    My 3rd year lectures on nonstationary processes for the academic year 2019-2020 are currently updated, after reviewing a long series of papers and books (over 60 and counting) on these issues for Mathematical Reviews®. The new section on deep learning and AI is expanding. The writing of the book supporting these lectures is in progress.
    List of lectures for the academic year 2018-2019:
    • Lecture 1 (Tuesday 10th January 2019): Volatility models. GARCH model: definition, properties, dependence structure, QML estimation. Slides
    • Lecture 2 (Friday 11th January 2019): Exponential GARCH model. Bootstrap methods and their application to model-free prediction and inference of univariate models. Multivariate models. Change-point models. Slides
    • Lecture 3 (Monday 25th February 2019): Long-range dependent and multifractal volatility models. Wavelet analysis of volatility processes. Slides
    • Lecture 4 (Tuesday 26th February 2019): Volatility estimation of high frequency time series. Deep Learning: convolutional, recurrent and hybrid architectures. Slides
  • 2009: Aarhus University, PhD lectures, based on the draft version of the book Large Sample Inference for Long Memory Processes (2012) Imperial College Press. In 2013, I wrote the review of this book for Mathematical Reviews®. If you are a MathSciNet subscriber, you could read this review from my author profile
    Additional material not covered in that book: Slides
  • 2006-2007: Ensae, lectures (in French) on Long-range dependence and change-points, Applications to univariate and multivariate financial time series,
    • Leçon 1 : Processus fortement dépendants, Transparents
    • Leçon 2 : Tests de détection de longue portée et estimateurs du paramètre de longue portée, Transparents
    • Leçon 3 : Tests de détection de ruptures, Transparents
    • Leçon 4 : Méthodes statistiques robustes aux ruptures, Transparents
    • Leçon 5 : Modèles multivariés, Transparents
  • 1993-1996: Lectures on optimization techniques with Maple. I have typed a 200 pages document, that obviously needs to be upgraded.

Programming Resources

  • The first class requires a good working knowledge of R
    Since the academic year 2016-2017, students undertaking their dissertation on Machine Learning, deep learning, etc., are using Python with the scikit-learn and TensorFlow libraries.
    The efficiency of the running of Python programs is substantially improved by calling them within a Julia program.
  • Some statistical procedures are using Octave, a matrix oriented programming language for numerical computing, the syntax of which is similar to Matlab, up to some changes in the syntax of a few functions.
  • The C++ compiler of the GNU Compiler Collection (GCC) is of great value. I'm using it since 1997. Valgrind is a useful tool for memory leak debugging, profiling, etc.
    NLopt is a free/open-source library for nonlinear optimization, callable from C, C++, Fortran, Python, Julia, R Octave, and Matlab programs.
  • Donald Knuth's home page. Don Knuth is the author of the celebrated books: The Art of Computer Programming You can find there everything which is important on semi-numerical and numerical algorithms, TeX etc.
  • The Digital Library of Mathematical Functions of the National Institute of Standards and Technology (NIST); See A special functions handbook for the digital age, R. Boisvert, C.W. Clark, D. Lozier, and F. Olver, Notices of the American Mathematical Society (2011), vol 58, 905-911, Pdf file. This new online library supersedes the celebrated book by M. Abramowitz and I. Stegun: the Handbook of Mathematical Functions
    The NIST Statistical Reference Datasets (NIST StRD) for assessing the numerical accuracy of statistical software packages. This database is very useful for checking parts of statistical procedures.
    Useful information on the issue of numerical accuracy of statistical software packages is available at B.D. McCullough's web page.
  • BOINC, the Berkeley Open Infrastructure for Network Computing, is an open-source software for volunteer and grid computing. BOINC projects are covering several fields: physics, astronomy, mathematics, biology, artificial intelligence, cryptography, computer science, etc. If your project is computationally very intensive, and intellectually attractive, you can transform it into a BOINC project.

Other Teaching and Supervision Resources

Update: April 2, 2019.