Weakly measurable function
In mathematics—specifically, in functional analysis—a weakly measurable function taking values in a Banach space is a function whose composition with any element of the dual space is a measurable function in the usual (strong) sense. For separable spaces, the notions of weak and strong measurability agree.
If (X, Σ) is a measurable space and B is a Banach space over a field K (usually the real numbers R or complex numbers C), then f : X → B is said to be weakly measurable if, for every continuous linear functional g : B → K, the function
is a measurable function with respect to Σ and the usual Borel σ-algebra on K.
A measurable function on a probability space is usually referred to as a random variable (or random vector if it takes values in a vector space such as the Banach space B). Thus, as a special case of the above definition, if (Ω, Σ, P) is a probability space, then a function Z: : Ω → B is called a (B-valued) weak random variable (or weak random vector) if, for every continuous linear functional g : B → K, the function
is a K-valued random variable (i.e. measurable function) in the usual sense, with respect to Σ and the usual Borel σ-algebra on K.
The relationship between measurability and weak measurability is given by the following result, known as Pettis' theorem or Pettis measurability theorem.
A function f is said to be almost surely separably valued (or essentially separably valued) if there exists a subset N ⊆ X with μ(N) = 0 such that f(X \ N) ⊆ B is separable.
Theorem (Pettis). A function f : X → B defined on a measure space (X, Σ, μ) and taking values in a Banach space B is (strongly) measurable (with respect to Σ and the Borel σ-algebra on B) if and only if it is both weakly measurable and almost surely separably valued.
In the case that B is separable, since any subset of a separable Banach space is itself separable, one can take N above to be empty, and it follows that the notions of weak and strong measurability agree when B is separable.