91Ó°ÊÓ

Brownian motion, also known as a Wiener process, is a fundamental concept in probability theory and mathematics. It is a continuous-time stochastic process \(W(t)\) that has several key properties:

• Starts at zero: \(W(0) = 0\).
• Has independent increments: the future path increments are independent of the past.
• The increments are normally distributed: \(W(t) - W(s)\) follows a normal distribution with mean zero and variance \(t - s\).
• Exhibits continuous paths: the graph is continuous with no jumps.

Brownian motion is used to model a wide range of real-world phenomena, from the stock market to diffusion processes in physics. It behaves like a fractal, zoomable at any scale, providing a rich structure for mathematical exploration. A critical aspect of Brownian motion is its use to model random fluctuations, making it a backbone of stochastic calculus.

The arcsine law is a fascinating result from probability theory, particularly relevant to Brownian motion. It describes a counter-intuitive property of the occupation time of a Brownian path:

• The time spent on one side of the origin is not uniformly distributed, even on symmetric paths.
• The distribution is heavily skewed toward extreme values, meaning that the path is more likely to spend a significant amount of time either entirely above or below the origin.

Mathematically, for a standard Brownian motion \(W(t)\), the arcsine law implies that the amount of time spent above zero in an interval \((0,1)\) follows:\[P\left(\frac{L_1}{1} \leq x\right) = \frac{2}{\pi} \arcsin\sqrt{x},\quad 0 \leq x \leq 1\]This result is counter-intuitive due to its bias toward extreme times, unlike what one might expect from a symmetric process. It exemplifies the surprising nature of processes like Brownian motion.

In probability and statistics, a density function describes the likelihood of a random variable to take on a given value. For continuous random variables, it is expressed with a probability density function (PDF), which provides the density of probability at each point.The density function derived in this exercise for \(U_x\) is:\[f_{U_x}(u) = \frac{1}{\pi \sqrt{u(1-u)}} \exp\left(-\frac{x^2}{2u}\right), ~~ 0 < u < 1\]Key features of density functions include the following:

• Non-negative over its domain.
• Integrates to one over the entire space, signifying total probability.
• Can be used to compute probabilities over intervals.

In this context, the density function describes the distribution of the time \(U_x\) spent under level \(x\) for a Wiener process in the window \((0,1)\). It incorporates the symmetry and probabilistic behavior of Brownian motion.

An indicator function is a simple yet powerful tool in mathematics and statistics, used to denote a conditional characteristic of a set or mathematical function:

• Expressed as \(I(A)\), it is 1 if event \(A\) occurs, and 0 otherwise.
• Used to transform sets and conditions into mathematical formulas.

In the context of this exercise, the indicator function \(I(W(t) < x)\) helps determine whether the Wiener process \(W(t)\) is below a fixed level \(x\) at time \(t\). This function is crucial to calculating \(U_x\), the time spent below this threshold. Indicator functions allow for setting up integrals that sum contributions where a condition holds over an interval, simplifying complex stochastic processes into more manageable forms. They form the basis for many operations in probability and are vital tools in measure theory.