This biography of a living person includes a list of general references, but it lacks sufficient corresponding inline citations. (July 2023) |
Howell Tong (simplified Chinese: 汤家豪; traditional Chinese: 湯家豪; pinyin: Tāng Jiāháo; born in 1944 in Hong Kong) is a statistician who has made fundamental contributions to nonlinear time series analysis, semi-parametric statistics, non-parametric statistics, dimension reduction, model selection, likelihood-free statistics and other areas. In the words of Professor Peter Whittle (FRS): "The striking feature of Howell Tong's … is the continuing freshness, boldness and spirit of enquiry which inform them-indeed, proper qualities for an explorer. He stands as the recognised innovator and authority in his subject, while remaining disarmingly direct and enthusiastic."[1] His work, in the words of Sir David Cox, "links two fascinating fields, nonlinear time series and deterministic dynamical systems."[2] He is the father of the threshold time series models, which have extensive applications in ecology, economics, epidemiology and finance. (See external links for detail.) Besides nonlinear time series analysis, he was the co-author of a seminal paper, which he read to the Royal Statistical Society, on dimension reduction in semi-parametric statistics by pioneering the approach based on minimum average variance estimation. He has also made numerous novel contributions to nonparametric statistics (obtaining the surprising result that cross-validation does not suffer from the curse of dimensionality for consistent estimation of the embedding dimension of a dynamical system), Markov chain modelling (with application to weather data), reliability, non-stationary time series analysis (in both the frequency domain and the time domain) and wavelets.