Tutorial, Bayes net for forensic DNA analysis, Satellite, Dice, DNA Helix, Distance Matrix of pairs of amino acids, Bristol Balloon festival
SuSTaIn About News Events Closing workshop... EC: High-dim Statistics... EC: Hidden Complexities Intractable Likelihoods UK Causal Inference Meeting EC: Astrostatistics Image Processing Extremes Structure and Uncertainty Functional data analysis Confronting Intractability Time for Causality Laurie Davies lectures Julian Besag memorial Research highlights Jobs Management Statistics Group Statistics Home Research Members Seminars Mathematics Home External Links APTS Complexity science Royal Statistical Society International Society for Bayesian Analysis


Bayesian wavelet estimators in nonparametric regression

Dr Natalia Bochkina (University of Edinburgh)

Web site: http://www.maths.ed.ac.uk/~nbochkin/

Dr Bochkina will present a series of 4 lectures, on 28-31 March 2011.

Note that the times vary from day to day.

  • Classical and Bayesian approaches to estimation in nonparametric regression, Monday 28 March, 3pm
  • Classical minimax consistency and concentration of posterior measures, Tuesday 29 March, 3pm
  • Wavelet estimators in nonparametric regression, Wednesday 30 March, 2pm
  • Wavelet estimators: simultaneous local and global optimality, Thursday 31 March, 11.10am

These should be of general interest, especially those whose work touches on Bayesian analysis, wavelets, or nonparametric regression. There are no pre-requisites for understanding the lectures, beyond say 3rd year undergraduate statistics.

Location: seminar room SM4, School of Mathematics, University of Bristol: Mathematics is building number 29 on the university precinct map.

Slides for lecture 1 | lecture 2 | lecture 3 | lecture 4


In the first two lectures I will give an overview of Bayesian and frequentist methods in nonparametric regression, such as kernel estimators, orthogonal basis estimators (Bayesian and non-Bayesian) and Bayesian estimators with Gaussian process priors (lecture 1), introduce the frameworks to study the validity of these methods, namely minimax consistency and concentration of posterior measures, and discuss such issues as the rate of convergence and adaptivity to unknown function smoothness (lecture 2).

The third lecture will be a review of Bayesian wavelet methods in nonparametric regression, and in the fourth lecture my latest work on simultaneous local and global optimality of Bayesian wavelet estimators will be presented.

Another wavelet seminar

Debashis Mondal (Chicago): Wavelet variance analysis for gappy time series data, 11am, Friday 1 April