Given a Borel map from a complete separable metric space into a separable metric space, every Borel probability measure has a conditional kernel given the map, unique up to a null set of the image measure; almost every conditional measure is carried by its fibre, and integrals against the measure are iterated integrals against the kernel and the image measure.
In the setting of Measure Spaces and the Lebesgue Integral: Standing Notation, let be a complete and separable metric space and a separable metric space, with Borel -algebras and ; let be measurable with respect to and ; let be a Borel measure on with , with image measure ; and let conditional kernels of given be those of The Conditional Kernel of a Probability Measure Given a Measurable Map §conditional-kernel, with . For the singleton is closed in , since for the open ball about of radius misses , hence belongs to by claim 1 of Borel Measurability and Bounded Integration on a Metric Space, so the fibre belongs to . For a measure and a function integrable with respect to it we also write for .
1. (Existence) There is a conditional kernel of given .
2. (Uniqueness) If and are conditional kernels of given , the set of the with belongs to and has -measure .
3. (Fibres) If is a conditional kernel of given , the set of the with belongs to and has -measure .
4. (Bounded functions) If is a conditional kernel of given and is measurable with respect to and bounded, then is integrable with respect to and to every , the function is measurable with respect to , bounded and integrable with respect to , and
5. (Nonnegative functions) Let be a conditional kernel of given and let be measurable with respect to , with . Then the set of the for which is not integrable with respect to belongs to , the function equal to for and to for is measurable with respect to , and is integrable with respect to if and only if and is integrable with respect to ; in that case .
Loading…
No relations recorded yet.