Problem 3 - A Course in Financial Calculus

Question

Suppose that \(W_{t}\) and \(\tilde{W}_{t}\) are independent Brownian motions under the measure \(\mathbb{P}\) and let \(\rho \in[-1,1]\) be a constant. Is the process \(X_{t}=\rho W_{t}+\sqrt{1-\rho^{2}} \tilde{W}_{t}\) a Brownian motion?

Step-by-Step Solution

Answer

Yes, \(X_t\) is a Brownian motion as it satisfies all the necessary properties.

Step 1: Define a Brownian Motion

A Brownian motion is a continuous-time stochastic process \(B_t\) with \(B_0 = 0\), for every \(s < t\), \(B_t - B_s\) is normally distributed with mean 0 and variance \(t-s\), and the increments \(B_t - B_s\) are independent of \(B_u\) for all \(u < s\).

Step 2: Express the Process in Terms of Known Brownian Motions

Given \(X_t = \rho W_t + \sqrt{1-\rho^2} \tilde{W}_t\), where \(W_t\) and \(\tilde{W}_t\) are independent Brownian motions, we need to determine if \(X_t\) satisfies the properties of a Brownian motion.

Step 3: Verify Initial Condition

Verify that \(X_0 = 0\), since \(W_0 = 0\) and \(\tilde{W}_0 = 0\) by the properties of Brownian motions, \(X_0 = \rho W_0 + \sqrt{1-\rho^2} \tilde{W}_0 = 0+0 = 0\).

Step 4: Check Normality and Variance of Increments

Consider \(X_t - X_s = \rho(W_t - W_s) + \sqrt{1-\rho^2}(\tilde{W}_t - \tilde{W}_s)\). Since \(W_t - W_s\) is normally distributed with mean 0 and variance \(t-s\), and similarly for \(\tilde{W}_t - \tilde{W}_s\), the linear combination \(X_t - X_s\) will also be normally distributed with mean 0. Compute the variance: \(\text{Var}(X_t - X_s) = \rho^2(t-s) + (1-\rho^2)(t-s) = (t-s)\).

Step 5: Verify Independence of Increments

Since \(W_t - W_s\) and \(\tilde{W}_t - \tilde{W}_s\) are independent, \(\rho(W_t - W_s) + \sqrt{1-\rho^2}(\tilde{W}_t - \tilde{W}_s)\) must also be independent of the past values \(X_u\) for \(u < s\).

Step 6: Conclude

The process \(X_t = \rho W_t + \sqrt{1-\rho^2} \tilde{W}_t\) satisfies all the properties of a Brownian motion: it starts at 0, has stationary and independent increments, and the increments are normally distributed with mean 0 and variance \(t-s\).

Key Concepts

Stochastic Process

A stochastic process is a collection of random variables representing the evolution of a system over time. These random variables are indexed by time, and they can model various phenomena where uncertainty and change are involved. For example, weather conditions, stock market fluctuations, or even physical phenomena like radioactive decay can be described by stochastic processes.

Some key features of stochastic processes include:

• Randomness: They incorporate an element of randomness, capturing the inherent unpredictability of the system.
• Time Indexing: The process evolves over time, with random variables assigned to each point in time.
• Dependence: The future states of the process can depend on its past states, revealing trends or patterns.

Brownian motion is a classic example of a continuous-time stochastic process, frequently used in finance and physics to model seemingly random movements.

Variance

Variance measures the spread or dispersion of a set of data points. In the context of a stochastic process like Brownian motion, variance plays a crucial role in determining how widely the process fluctuates over time.

For Brownian motion, the variance of increments over a time interval \(t-s\) is given by \(t-s\), meaning variance grows linearly with time:

• A larger variance suggests more variability or spread in the data points.
• Calculating the variance helps predict the magnitude of future deviations of the process.

Understanding variance is vital for interpreting results in fields like finance, where it can indicate the risk or volatility of an investment over time. A stable or predictable variance can also provide insights into the consistency of a given process or data set.

Independent Increments

A process exhibits independent increments if the increments are statistically independent from one another. This means that the change in the process over any interval is independent of the change over non-overlapping intervals. For Brownian motion, this property of independent increments is fundamental.

Some important aspects to remember include:

• Independence implies that the occurrence of one event does not affect the probability of another.
• Independent increments suggest no memory effect; the process starts anew at each interval without influence from previous changes.
• This independence allows for simplified calculations and modeling of processes.

In the context of a Brownian motion process, the independent increments ensure that the changes in position from time point to time point are unpredictable, capturing the essence of true random movement.

Normal Distribution

A normal distribution, often referred to as a Gaussian distribution, is a probability distribution that is symmetric about the mean, depicting that the data near the mean are more frequent in occurrence than data far from the mean.

Here are some key points about normal distribution in the context of Brownian motion:

• The increments of Brownian motion follow a normal distribution with mean 0, reflecting that the average change at any point is zero.
• Understanding normal distribution is crucial because it provides a basis for many statistical methods and theories.
• The shape of the normal distribution curve, often described as a "bell curve," is determined by two parameters: the mean and the variance.

In scientific and financial applications, the normal distribution helps model random variables and is used extensively to describe processes that naturally oscillate around a steady state, like the stock prices or physical diffusion processes.