Research project

Confidence Bands For Linear Regression - W Lui

Project overview

Regression analysis is one of the most commonly employed statistical techniques. It has applications in all scientific areas, such as engineering, physics, chemistry, biology, finance, and medicine. Simultaneous confidence bands for regression models can be used not only to assess the plausible range of a regression model but also to gain quantitative information about the possible differences between different regression models. So the methodology of simultaneous confidence bands is an integral part of regression analysis and has immediate applications to data analysis.The proposed research is to study the construction of optimal confidence band which provides the most accurate information about a regression model under certain recently proposed optimality criterion, and to study ways of constructing simultaneous confidence bands of exact 1-a level for assessing the all-contrast differences of several regression model under some very general setting.

Staff

Lead researchers

Professor Wei Liu

Professor in Mathematics
Research interests
  • Simultaneous inference (including multiple tests and simultaneous confidence bands);
  • Simultaneous tolerance bands and statistical calibration;
  • Tolerance bounds, tolerance regions and applications;
Connect with Wei

Collaborating research institutes, centres and groups

Research outputs

Mortaza Jamshidian, Wei Liu & Frank Bretz, 2010, Computational Statistics and Data Analysis, 54(6), 1475-1483
Type: article
Wei Liu & Pascal Ah-kine, 2010, Journal of Statistical Planning and Inference, 140(5), 1225-1235
Type: article
A.J. Hayter, W. Liu & P. Ah-Kine, 2009, Journal of Statistical Planning and Inference, 139(2), 329-334
Type: article