Speaker: Chenlei Leng (Warwick)
Title: Least squares in high-dimensional linear models
Abstract: Ordinary least squares (OLS) is the default method for fitting linear models, but is not applicable for high-dimensional problems. This talk re-examines OLS and advocates two novel algorithms that only require least-squares fitting. Ongoing work that aims at consistent model selection under general conditions in a penalised least-squares setting will also be discussed.