(Properties of Conditional Expectation). (a) Tower Property....

(Properties of Conditional Expectation). (a) Tower Property. Use the definition of conditional expectation to prove the tower property: If $\mathcal{H}$ is a subfield of $\mathcal{G}$, then, $$ E(E(X \mid G) \mid \mathcal{H})=E(X \mid \mathcal{H}) $$ (b) Factorization Property. If $Y \in \mathcal{G} \subset$, show that if $E(|X|)<\infty$ and $Y$ is bounded then $$ E(X Y \mid G)=Y E(X \mid G) $$ A good part of the challenge of problems like these is the introduction of notation that makes clear that there has been no "begging of the question." Without due care, one can easily make accidental use of the tower property to prove the tower, property.

(Properties of Conditional Expectation).
(a) Tower Property. Use the definition of conditional expectation to prove the tower property: If $\mathcal{H}$ is a subfield of $\mathcal{G}$, then,
$$
E(E(X \mid G) \mid \mathcal{H})=E(X \mid \mathcal{H})
$$
(b) Factorization Property. If $Y \in \mathcal{G} \subset$, show that if $E(|X|)<\infty$ and $Y$ is bounded then
$$
E(X Y \mid G)=Y E(X \mid G)
$$
A good part of the challenge of problems like these is the introduction of notation that makes clear that there has been no "begging of the question." Without due care, one can easily make accidental use of the tower property to prove the tower, property.

Key Concepts

Conditional Expectation

This concept refers to the expected value of a random variable given a particular sigma?algebra. It is defined such that it becomes a measurable function with respect to the given sigma?algebra, and it satisfies the property that its integral over any measurable set in the sigma?algebra is equal to the integral of the original random variable over that set. This forms the basis for many properties and operations in probability theory that exploit conditional independence and measurability.

Sigma-Algebras (Subfields)

Sigma-algebras are collections of sets which satisfy certain axioms (closure under countable unions and complements) and provide the formal framework for measurability. When considering conditional expectation, sub-sigma-algebras (or subfields) are used to express the information available, and the ‘conditional’ aspect relates to being measurable with respect to one of these sigma-algebras. The relationship between sigma-algebras is crucial when proving properties like the tower property.

Tower Property

Also known as the law of iterated expectations, the tower property states that taking the conditional expectation iteratively with respect to nested sigma-algebras is equivalent to taking a single conditional expectation with respect to the smallest sigma-algebra. This property is fundamental in simplifying complex conditional expectations and ensuring that no additional assumptions are made inadvertently during proofs.

Factorization Property

This property involves the interaction between conditional expectation and multiplication by bounded or measurable functions. Specifically, if a random variable is measurable with respect to a given sigma-algebra, then it can be factored out of the conditional expectation, effectively turning a conditional expectation of a product into the product of the measurable function and the conditional expectation of the other factor. This is an important tool in manipulating and simplifying expressions involving conditional expectations in probability theory.

Transcript

Please give Ace some feedback