| Adrian Baddeley | |
| Birth Place: | Melbourne, Australia |
| Field: | Statistics |
| Education: | Australian National University University of Cambridge |
| Doctoral Advisor: | David George Kendall |
| Workplaces: | University of Bath University of Western Australia CSIRO Curtin University |
| Awards: |
|
Adrian John Baddeley (born May 25, 1955)[1] is a statistical scientist working in the fields of spatial statistics,[2] statistical computing, stereology[3] and stochastic geometry.
Baddeley was born in Melbourne, Australia and educated at Eltham High School there, and studied mathematics and statistics at the Australian National University (honours supervisor: Roger Miles) and the University of Cambridge (PhD supervisor: David George Kendall). He was elected a Junior Research Fellow at Trinity College, Cambridge in the second year of his PhD. Subsequently, he worked for the University of Bath (1982–85), the CSIRO Division of Mathematics and Statistics, Sydney (1985–88), the Centrum Wiskunde & Informatica, Amsterdam, the Netherlands (1988–94), the University of Western Australia (where he was Professor of Statistics from 1994 to 2010), CSIRO Division of Mathematics, Informatics and Statistics, Perth (2010-2012), and the Centre for Exploration Targeting at the University of Western Australia (2013-2014). He is now Professor of Computational Statistics at Curtin University.
Classical methods of stereology were limited by the requirement that the cutting plane be randomly oriented. Baddeley developed an alternative technique[4] in which the cutting plane is "vertical" (parallel to a fixed axis, or perpendicular to a fixed surface) making it possible to apply quantitative microscopy to cylindrical core samples, samples of flat materials, and longitudinal sections.
Baddeley is a leading advocate of statistical ideas in stereology. With Cruz-Orive he demonstrated the role of the Horvitz-Thompson weighting principle and the Rao-Blackwell theorem in stereological sampling.
Baddeley is one of the world leading specialists in point pattern analysis, a connection of stochastics and geometry applied to the analysis of (mainly) 2D point distributions in euclidean space. He has developed statistical methodologies for analyzing the structure of spatial patterns of points, including methods based on survival analysis,[5] nonparametrics,[6] [7] new point process models,[8] [9] model-fitting principles (i.e. 'regression analysis' for point patterns) and algorithms[10] [11] [12] and open-source software.[13]