A. García Nogales, P. Pérez Fernández
A classic Phillips result on the relationship between independence and conditional
independence appears here as a particular case of a result recently published by
the authors. Specifically, this result establishes that, given random variables X, Y,Z and
under the conditional independence of X and Y given Z, the independence of X and Y
is equivalent to the independence (as Markov kernels) of the conditional distributions of
X given Z and of Y given Z.
Keywords: Conditional independence, Markov kernel
Scheduled
Posters I
June 7, 2022 12:00 PM
Faculty Hall