Dr. W. van Ackooij (Electricité de France R&D, Paris): 
<p>
In this talk we will consider probability constraints acting on
nonlinear systems involving an elliptically symmetrically distributed
random vector. The nonlinear system is assumed to be convex in the decision
vector. Under these assumptions we can provide a series of tools that allow
us to establish novel convexity results for upper level sets of the
probability function provided a sufficiently large threshold is taken. This
latter threshold can be computed analytically.