Justin Wishart
Department of Statistics
Macquarie University
Australia
Biography
My research is situated in the classical and somewhat general field of nonparametric regression in Mathematical Statistics. I use nonparametric regression estimation techniques including wavelets kernel smoothing and splines to identify key features of data. This includes change point feature estimation (sharp changes in responses or change in trends). Another useful technique is dimension reduction which takes a high dimension dataset and maps it to a lower dimensional dataset to aid in creating more powerful regression estimates, this is implemented in a single index framework. My current work focuses primarily on nonparametric inverse problems which a particular focus on deconvolution in anisotropic and isotropic random fields.
Research Interest
1) Provide data-driven automatic tools for nonparametric deconvolution regression models. 2) Extend nonparamteric regression methodology to higher dimensional data with fractal properties. 3) Provide useful statistical software to the scientific community.
Publications
-
Wishart JR. Wavelet deconvolution in a periodic setting with long-range dependent errors. Journal of Statistical Planning and Inference. 2013 May 31;143(5):867-81.
-
Wishart J. Kink estimation with correlated noise. Journal of the Korean Statistical Society. 2009 Jun 30;38(2):131-43.
-
Scalbert A, Brennan L, Manach C, Andres-Lacueva C, Dragsted LO, Draper J, Rappaport SM, van der Hooft JJ, Wishart DS. The food metabolome: a window over dietary exposure. The American journal of clinical nutrition. 2014 Jun 1;99(6):1286-308.
