Statistical Inference in Complex Problems: Specifying Real Uncertainty
ABSTRACT:
This talk is intended for statisticians who will be facing real
challenges involving complex data. It summarizes statistical methods
used in recent applied problems, mainly related to health. The methods
discussed include multivariate methods, hierarchical models, repeated
measure models and general regression methods. A number of guiding
principles for data analysis are set out and illustrated by the
examples.
About the Speaker
David Andrews is professor emeritus in the department of statistics
at the University of Toronto. Prior to his retirement at the end of
2001, Professor Andrews also held a joint position in the Department of
Public Health Sciences at University of Toronto. His interests span the
range of applied statistics and modelling of complex data. Among his
recent interests is statistical genetics. He has also written a book
(with J. Stafford) on symbolic computation in statistics.
