The Lebesgue measure plays a fundamental role in stochastic calculus, as it is the standard measure on the underlying probability space that represents the randomness of the stochastic processes involved. More specifically, stochastic calculus deals with the study of stochastic processes and their properties, such as Brownian motion, stochastic integrals, and stochastic differential equations. The Lebesgue measure is used to describe the nature of their sample paths and to define the integral of these processes with respect to time.
To understand the significance of the Lebesgue measure in stochastic calculus, it is essential to establish a framework for the probability space (Ω, ℱ, ℙ), where Ω represents the sample space, ℱ denotes the σ-algebra of events, and ℙ corresponds to the probability measure. In this probability space, ℙ is required to satisfy the axioms of probability, and the Lebesgue measure serves as its foundation.
The relationship between the Lebesgue measure and the Radon-Nikodym derivative lies within the concept of absolutely continuous measures. Let μ and ν be two measures on the measurable space (Ω, ℱ). We say that ν is absolutely continuous with respect to μ if for any event A ∈ ℱ, μ(A) = 0 implies ν(A) = 0. In notation, we write
ν ≪ μ.
The Radon-Nikodym theorem establishes a connection between absolutely continuous measures, Lebesgue measure, and the Radon-Nikodym derivative. It states that if ν is absolutely continuous with respect to μ, there exists a non-negative measurable function f (unique up to a μ-null set) such that for every A ∈ ℱ,
ν(A) = ∫Af dμ.
The function f is called the Radon-Nikodym derivative of ν with respect to μ, which we denote as $f = \frac{\mathrm{d}\nu}{\mathrm{d}\mu}$. The Radon-Nikodym derivative characterizes the rate of change of one measure (in this case, the Lebesgue measure) with respect to another. In the context of stochastic calculus, the Radon-Nikodym derivative becomes particularly relevant when finding the change of measure or in Girsanov’s theorem, which enables the transformation of a given stochastic process under a new probability measure, simplifying the analysis or numerical computations.
To illustrate the importance of the Lebesgue measure and the Radon-Nikodym derivative in stochastic calculus, consider the following example. Let Wt be a standard Brownian motion. We want to construct a new probability measure, denoted by ℚ, under which the Brownian motion has an additional drift term, i.e., Wtℚ = Wt + λt for some constant λ. The Radon-Nikodym derivative will be instrumental in finding the proper change of measure.
Applying Girsanov’s theorem, we can find the Radon-Nikodym derivative of ℚ with respect to ℙ as
$$\frac{\mathrm{d}\mathbb{Q}}{\mathrm{d}\mathbb{P}} = \exp \left( -\lambda W_t - \frac{1}{2} \lambda^2 t \right),$$
where Wt is the standard Brownian motion under the original Lebesgue measure-induced probability, ℙ. With the Radon-Nikodym derivative, we can compute expectations or evaluate integrals with respect to either measure, ℙ or ℚ, thereby relating the quantities of interest under both measures.
In conclusion, the Lebesgue measure is a fundamental building block in stochastic calculus, providing the foundation for the probability measure on the underlying space. The Radon-Nikodym derivative allows us to relate different probability measures in the study of stochastic processes and is a crucial tool for changing measures and analyzing processes under alternative probability measures.