Fast Approximate Inference Methods for Flexible Regression
Funding: 2011: $110,000
Project Member(s): Wand, M.
Funding or Partner Organisation: Australian Research Council (ARC Discovery Projects)
Start year: 2011
Summary: Flexible regression is now a mainstay of statistical analysis of complex data sets. This project will gear up flexible regression for larger problems brought about by the current era of rapid technological change. It will draw upon approximate inference methods developed in Computer Science, leading to new statistical methodology for fast analyses. It will also develop novel statistical theory and new methodology for neuroimage data. The project will allow analyses of large data sets that would not even be attempted using currently available methods, opening up a wide range of new applications.
Wand, M, Ormerod, JT & Pham, T 2013, 'Mean field variational Bayesian inference for nonparametric regression with measurement error', Computational Statistics and Data Analysis, vol. 68, no. 1, pp. 375-387.
View/Download from: UTS OPUS or Publisher's site
Keywords: Variational Bayes, Penalised splines, wavelets, expectation propagation, longitudinal data analysis, measurement error models, neuroscience.
FOR Codes: Statistical Theory, Applied Statistics, Expanding Knowledge in the Mathematical Sciences, Expanding Knowledge in the Medical and Health Sciences, Statistics not elsewhere classified