Teaching


Linear models and analysis of variance

  • Particulars: University College London, 1st term, 2010-2011, 2011-2012, 2012-2013, 2013-2014, 2nd year undergraduate.
  • Module codes: STAT2002, STAT7101
  • Description: The aim of this course is to provide an introduction to linear statistical modelling and to the analysis of variance with emphasis on ideas, methods, applications and interpretation of results. On successful completion of the course, a student will have an understanding of the basic ideas underlying multiple regression and the analysis of variance. Furthermore, the student will be able to analyse, using a statistical package, data from some common experimental layouts, and understand and know how to check the validity of the assumptions underlying such analyses.
  • Resources: available at UCL Moodle (restricted access).

Introduction to practical Statistics

  • Particulars: University College London, 2nd term, 2010-2011, 1st year undergraduate.
  • Module code: STAT1006
  • Description: The aim of this course is to provide an accessible and application-oriented introduction to basic ideas in probability and statistics. On successful completion of the course, a student should be able to identify and carry out appropriate statistical analyses of much-encoutered data settings using a statistical software package and interpret the resultant output.
  • Resources: available at UCL Moodle (restricted access).

Advanced topics in Statistics: Asymptotic Statistics

  • Particulars: University of Warwick, 1st term, 2009-2010, 4th year undergraduate and MSc.
  • Module code: ST414
  • Description: The aim of this topics course is to introduce students to basic asymptotic methods that are used in Statistics and to highlight their importance. Special attention is paid to likelihood based asymptotics for parametric models, focusing on the asymptotic properties of the maximum likelihood estimator. The course combines the development of general asymptotic tools with their application to some much-used results in statistics (such as, for example, the derivation of the asymptotic distribution of the log likelihood-ratio statistic, the 1/2 adjustment to the binomial counts for bias reduction in log-odds estimation etc.). Advantages and shortcomings of different asymptotic methods are explored using computer simulation.
  • Resources: available here zip.