Grace L. Yang
Department of Mathematics and Statistics, University of Maryland
on a 
field 
and 
on a field 
. A fundamental concept
in Le Cam's asymptotic theory is to
approximate a statistical experiment by a more tractable experiment.
The approximation is made precise in terms of
the Le Cam distance 
between two experiments,
and 
. In this presentation, we highlight
the approximations that yield local and global asymptotic normality.
The local theory leads to the well known
Hájek-Le Cam asymptotic minimax theorem and the Hájek-Le Cam
convolution theorem. Recent results of Brown and Low, and Nussbaum show
that many regression problems or problems of density estimation
in the i.i.d. case are asymptotic normal in the global sense.