STA 376: Advanced Modelling & Scientific Computing
- Spring 2006 -

STA 356 Schedule Support



  • There is no course text. Many statistics texts (and of course web sites - go and google) include discussion - often in useful context - of standard numerical methods we'll draw on. Some to note will be added here:

    • Much useful and relevant material on modelling and computation in statistics appears throughout the key text Bayesian Data Analysis by Gelman, Meng, Stern, and Rubin (Chapman and Hall, 2003, 2nd ed.). Most of you will already be very familiar with the book. We will repeatedly refer to sections of this book in the course.
    • K. Lange, Numerical Analysis for Statisticians (Springer Verlag, 1998) is a very good, modern reference text on numerical methods. (numerical quadrature for one or low-dimensional numerical integration, gradient-based search/mode hunting, the EM algorithm, aspects of Monte Carlo simulation, asymptotic approximations including Laplace approximations, and others that are just basic in statistical modelling. The book also contains useful coverage of many other topics (transforms, matrix algebra, nonlinear equations, etc) and provides a good general reference.

    • General references:
      • Bayesian Theory, Bernardo and Smith, 1994, Wiley.
      • Numerical Recipes in C, Press, Tuelosky, Vetterling and Flannery, 1992 (2nd Edn), Cambridge University Press.
      • Handbook of Mathematical Functions, eds: Abramowitz and Stegun, 1970, Dover.

  • STA 214 notes from my STA 214 class, that includes a great deal of development of basic ideas, theory and methods of simulation-based computation (Monte Carlo and MCMC) in statistical modelling -- material that should be stock knowledge and is pre-requisite for STA 376.

    The STA 214 site also contains much more on the Schedule and Support pages - including a heap of (potentially) useful Matlab code and examples.