Abstract:
The goal of this talk will be to analyze important
properties of distributions in terms of their moments.
A lot of attention will be paid to conditions under which
a distribution is, or is not, uniquely determined by the
moment sequence. This will be related to characterizing
the distributions of non-linear (Box-Cox) transformations
of random data. Besides well-known classical results, several
recent results will be reported. The results will be well-illustrated
by examples involving popular distributions. Some of them are new, not
so well-known, and look even shocking. If time allows,
a few open questions will be outlined.
The talk will be addressed not only to professionals in probability
and statistics but also to Master and Doctoral students in this area.
|
