Financial Mathematics Unit 3 ReviewProbability Theory & Random Variables

Key Concepts and Definitions

• Probability measures the likelihood of an event occurring and ranges from 0 (impossible) to 1 (certain)
• Sample space ($\Omega$) includes all possible outcomes of an experiment or random process
• Event (A) is a subset of the sample space containing one or more outcomes of interest
• Mutually exclusive events cannot occur simultaneously in a single trial (rolling a 3 and a 4 on a die)
• Independent events do not influence each other's probability (flipping a coin and rolling a die)
  • $P(A \cap B) = P(A) \cdot P(B)$ for independent events A and B
• Conditional probability $P(A|B)$ measures the probability of event A given that event B has occurred
  • Calculated using the formula $P(A|B) = \frac{P(A \cap B)}{P(B)}$

Probability Basics

• Addition rule for mutually exclusive events: $P(A \cup B) = P(A) + P(B)$
• Addition rule for non-mutually exclusive events: $P(A \cup B) = P(A) + P(B) - P(A \cap B)$
• Multiplication rule for independent events: $P(A \cap B) = P(A) \cdot P(B)$
• Multiplication rule for dependent events: $P(A \cap B) = P(A) \cdot P(B|A)$
• Complementary events have probabilities that sum to 1 (rolling an even number and rolling an odd number on a die)
  • $P(A) + P(A^c) = 1$, where $A^c$ is the complement of event A
• Bayes' theorem relates conditional probabilities: $P(A|B) = \frac{P(B|A) \cdot P(A)}{P(B)}$

Random Variables

• Random variable (X) assigns a numerical value to each outcome in a sample space
• Discrete random variables have countable outcomes (number of heads in 5 coin flips)
• Continuous random variables have uncountable outcomes within an interval (time until a stock reaches a certain price)
• Probability mass function (PMF) gives the probability of each value for a discrete random variable
  • $p(x) = P(X = x)$, where x is a possible value of X
• Probability density function (PDF) describes the probability distribution for a continuous random variable
  • $f(x)$ is the PDF, and $P(a \leq X \leq b) = \int_a^b f(x) dx$

Probability Distributions

• Bernoulli distribution models a single trial with two possible outcomes (success or failure)
  • $P(X = 1) = p$ and $P(X = 0) = 1 - p$, where p is the probability of success
• Binomial distribution models the number of successes in a fixed number of independent Bernoulli trials
  • $P(X = k) = \binom{n}{k} p^k (1-p)^{n-k}$, where n is the number of trials and k is the number of successes
• Poisson distribution models the number of events occurring in a fixed interval of time or space
  • $P(X = k) = \frac{e^{-\lambda} \lambda^k}{k!}$, where $\lambda$ is the average rate of events
• Normal (Gaussian) distribution is a continuous distribution with a bell-shaped curve
  • Characterized by its mean ($\mu$) and standard deviation ($\sigma$)
  • Standard normal distribution has $\mu = 0$ and $\sigma = 1$

Expectation and Variance

• Expected value (mean) of a random variable is the average value over many trials
  • For discrete X: $E(X) = \sum_{x} x \cdot p(x)$
  • For continuous X: $E(X) = \int_{-\infty}^{\infty} x \cdot f(x) dx$
• Variance measures the average squared deviation from the mean
  • $Var(X) = E[(X - \mu)^2] = E(X^2) - [E(X)]^2$
• Standard deviation is the square root of the variance: $\sigma = \sqrt{Var(X)}$
• Linearity of expectation: $E(aX + bY) = aE(X) + bE(Y)$ for constants a and b
• Covariance measures the linear relationship between two random variables X and Y
  • $Cov(X, Y) = E[(X - E(X))(Y - E(Y))]$

Applications in Finance

• Portfolio return is a weighted average of individual asset returns
  • $R_p = \sum_{i=1}^n w_i R_i$, where $w_i$ is the weight of asset i and $R_i$ is its return
• Portfolio variance depends on asset variances and covariances
  • $\sigma_p^2 = \sum_{i=1}^n w_i^2 \sigma_i^2 + 2 \sum_{i=1}^n \sum_{j=i+1}^n w_i w_j \sigma_i \sigma_j \rho_{ij}$
  • $\rho_{ij}$ is the correlation coefficient between assets i and j
• Value at Risk (VaR) estimates the potential loss for an investment over a given time horizon and confidence level
  • For a normal distribution: $VaR = \mu + z_{\alpha} \sigma$, where $z_{\alpha}$ is the z-score for the desired confidence level
• Option pricing models (Black-Scholes) use probability distributions to estimate the fair price of options contracts

Common Probability Problems

• Gambler's Ruin problem analyzes the probability of a gambler going bankrupt
  • Depends on initial capital, win probability, and bet size
• Birthday Problem calculates the probability of two people sharing a birthday in a group
  • Surprisingly high for relatively small groups (50% for 23 people)
• Monty Hall problem demonstrates the importance of conditional probability
  • Switching doors after the host reveals a goat increases the win probability to 2/3
• Coupon Collector's problem determines the expected number of trials to collect all unique items
  • $E(X) = n(\frac{1}{n} + \frac{1}{n-1} + ... + \frac{1}{2} + 1)$, where n is the number of unique items

Key Formulas and Techniques

• Combinatorics for counting possibilities
  • Permutations: $nPr = \frac{n!}{(n-r)!}$
  • Combinations: $nCr = \binom{n}{r} = \frac{n!}{r!(n-r)!}$
• Moment generating functions (MGFs) uniquely characterize probability distributions
  • $M_X(t) = E(e^{tX}) = \sum_{x} e^{tx} \cdot p(x)$ for discrete X
  • $M_X(t) = E(e^{tX}) = \int_{-\infty}^{\infty} e^{tx} \cdot f(x) dx$ for continuous X
• Law of Large Numbers states that the sample mean approaches the population mean as the sample size increases
  • $\lim_{n \to \infty} P(|\bar{X}_n - \mu| < \epsilon) = 1$ for any $\epsilon > 0$
• Central Limit Theorem asserts that the sum of many independent random variables is approximately normally distributed
  • Useful for approximating probabilities for large samples
• Markov Chains model systems transitioning between states with fixed probabilities
  • Transition matrix P contains the probabilities of moving from one state to another