📈Theoretical Statistics Unit 1 – Probability Theory Basics

Key Concepts and Definitions

• Probability measures the likelihood of an event occurring ranges from 0 (impossible) to 1 (certain)
• Sample space ($\Omega$) set of all possible outcomes of a random experiment
• Event (E) subset of the sample space represents a specific outcome or set of outcomes
• Mutually exclusive events cannot occur simultaneously in a single trial (rolling a 1 and a 2 on a fair die)
• Collectively exhaustive events cover all possible outcomes in the sample space
  • Example: rolling a number less than 4 and rolling a number greater than or equal to 4 on a 6-sided die
• Complement of an event ($E^c$) consists of all outcomes in the sample space that are not in the event E
• Union of events ($E \cup F$) includes all outcomes that are in either event E or event F, or both
• Intersection of events ($E \cap F$) includes only the outcomes that are common to both events E and F

Sample Spaces and Events

• Discrete sample space has a finite or countably infinite number of possible outcomes (rolling a die, flipping a coin)
• Continuous sample space has an uncountably infinite number of possible outcomes (measuring the height of a person)
• Simple event consists of a single outcome from the sample space (drawing a specific card from a deck)
• Compound event combines two or more simple events (drawing a red card and then a black card from a deck)
• Venn diagrams visually represent relationships between events using overlapping circles or other shapes
  • Example: two overlapping circles representing events A and B, with the overlapping region representing $A \cap B$
• Tree diagrams illustrate all possible outcomes of a sequence of events using branches and nodes
• Permutations count the number of ways to arrange a set of objects in a specific order
• Combinations count the number of ways to select a subset of objects from a larger set, disregarding the order

Probability Axioms and Rules

• Non-negativity axiom: $P(E) \geq 0$ for any event E in the sample space
• Normalization axiom: $P(\Omega) = 1$, where $\Omega$ is the entire sample space
• Additivity axiom: for mutually exclusive events $E_1, E_2, \ldots$, $P(\bigcup_{i=1}^{\infty} E_i) = \sum_{i=1}^{\infty} P(E_i)$
• Complement rule: $P(E^c) = 1 - P(E)$, where $E^c$ is the complement of event E
• Addition rule: $P(E \cup F) = P(E) + P(F) - P(E \cap F)$ for any two events E and F
  • Simplifies to $P(E \cup F) = P(E) + P(F)$ when E and F are mutually exclusive
• Multiplication rule: $P(E \cap F) = P(E) \cdot P(F|E)$, where $P(F|E)$ is the conditional probability of F given E
  • Simplifies to $P(E \cap F) = P(E) \cdot P(F)$ when E and F are independent events
• Inclusion-exclusion principle calculates the probability of the union of multiple events by considering their intersections

Conditional Probability and Independence

• Conditional probability $P(F|E)$ measures the probability of event F occurring given that event E has already occurred
  • Formula: $P(F|E) = \frac{P(E \cap F)}{P(E)}$, where $P(E) > 0$
• Independence two events E and F are independent if the occurrence of one does not affect the probability of the other
  • Mathematically, $P(E \cap F) = P(E) \cdot P(F)$ or equivalently, $P(F|E) = P(F)$ and $P(E|F) = P(E)$
• Bayes' theorem relates conditional probabilities $P(E|F)$ and $P(F|E)$
  • Formula: $P(E|F) = \frac{P(F|E) \cdot P(E)}{P(F)}$, where $P(F) > 0$
• Law of total probability expresses the probability of an event as a sum of conditional probabilities
  • Formula: $P(F) = \sum_{i=1}^{n} P(F|E_i) \cdot P(E_i)$, where $E_1, E_2, \ldots, E_n$ form a partition of the sample space
• Chain rule (multiplication rule for conditional probabilities) calculates the probability of the intersection of multiple events
  • Formula: $P(E_1 \cap E_2 \cap \ldots \cap E_n) = P(E_1) \cdot P(E_2|E_1) \cdot P(E_3|E_1 \cap E_2) \cdot \ldots \cdot P(E_n|E_1 \cap E_2 \cap \ldots \cap E_{n-1})$

Random Variables and Distributions

• Random variable (X) function that assigns a real number to each outcome in a sample space
• Discrete random variable has a countable number of possible values (number of heads in 10 coin flips)
• Continuous random variable has an uncountable number of possible values within a range (time taken to complete a task)
• Probability mass function (PMF) for a discrete random variable X, denoted by $p_X(x)$, gives the probability of X taking on a specific value x
  • Properties: $p_X(x) \geq 0$ for all x, and $\sum_x p_X(x) = 1$
• Cumulative distribution function (CDF) for a random variable X, denoted by $F_X(x)$, gives the probability of X being less than or equal to a specific value x
  • Formula: $F_X(x) = P(X \leq x)$
  • Properties: $0 \leq F_X(x) \leq 1$, $\lim_{x \to -\infty} F_X(x) = 0$, $\lim_{x \to \infty} F_X(x) = 1$, and $F_X(x)$ is non-decreasing
• Probability density function (PDF) for a continuous random variable X, denoted by $f_X(x)$, is used to calculate probabilities for ranges of values
  • Properties: $f_X(x) \geq 0$ for all x, and $\int_{-\infty}^{\infty} f_X(x) dx = 1$
  • Relationship with CDF: $F_X(x) = \int_{-\infty}^{x} f_X(t) dt$

Expectation and Variance

• Expectation (mean) of a discrete random variable X, denoted by $E[X]$ or $\mu_X$, is the weighted average of all possible values
  • Formula: $E[X] = \sum_x x \cdot p_X(x)$
• Expectation of a continuous random variable X is calculated using the PDF
  • Formula: $E[X] = \int_{-\infty}^{\infty} x \cdot f_X(x) dx$
• Linearity of expectation for random variables X and Y and constants a and b: $E[aX + bY] = aE[X] + bE[Y]$
• Variance of a random variable X, denoted by $Var(X)$ or $\sigma_X^2$, measures the average squared deviation from the mean
  • Formula for discrete X: $Var(X) = E[(X - \mu_X)^2] = \sum_x (x - \mu_X)^2 \cdot p_X(x)$
  • Formula for continuous X: $Var(X) = \int_{-\infty}^{\infty} (x - \mu_X)^2 \cdot f_X(x) dx$
• Standard deviation $\sigma_X$ is the square root of the variance
• Properties of variance: $Var(aX + b) = a^2Var(X)$ for constants a and b, and $Var(X + Y) = Var(X) + Var(Y)$ for independent random variables X and Y
• Covariance $Cov(X, Y)$ measures the linear relationship between two random variables X and Y
  • Formula: $Cov(X, Y) = E[(X - \mu_X)(Y - \mu_Y)]$
• Correlation coefficient $\rho_{X,Y}$ standardizes covariance to be between -1 and 1
  • Formula: $\rho_{X,Y} = \frac{Cov(X, Y)}{\sigma_X \sigma_Y}$

Common Probability Distributions

• Bernoulli distribution models a single trial with two possible outcomes (success with probability p, failure with probability 1-p)
  • PMF: $p_X(x) = p^x (1-p)^{1-x}$ for $x \in \{0, 1\}$
  • Mean: $E[X] = p$, Variance: $Var(X) = p(1-p)$
• Binomial distribution models the number of successes in a fixed number of independent Bernoulli trials
  • PMF: $p_X(x) = \binom{n}{x} p^x (1-p)^{n-x}$ for $x \in \{0, 1, \ldots, n\}$
  • Mean: $E[X] = np$, Variance: $Var(X) = np(1-p)$
• Poisson distribution models the number of rare events occurring in a fixed interval of time or space
  • PMF: $p_X(x) = \frac{e^{-\lambda} \lambda^x}{x!}$ for $x \in \{0, 1, 2, \ldots\}$
  • Mean: $E[X] = \lambda$, Variance: $Var(X) = \lambda$
• Uniform distribution models a random variable with constant probability density over a specified range
  • PDF (continuous): $f_X(x) = \frac{1}{b-a}$ for $x \in [a, b]$
  • Mean: $E[X] = \frac{a+b}{2}$, Variance: $Var(X) = \frac{(b-a)^2}{12}$
• Normal (Gaussian) distribution models many natural phenomena and has a bell-shaped PDF
  • PDF: $f_X(x) = \frac{1}{\sigma \sqrt{2\pi}} e^{-\frac{(x-\mu)^2}{2\sigma^2}}$ for $x \in (-\infty, \infty)$
  • Mean: $E[X] = \mu$, Variance: $Var(X) = \sigma^2$
• Exponential distribution models the time between rare events in a Poisson process
  • PDF: $f_X(x) = \lambda e^{-\lambda x}$ for $x \geq 0$
  • Mean: $E[X] = \frac{1}{\lambda}$, Variance: $Var(X) = \frac{1}{\lambda^2}$

Applications and Problem-Solving Techniques

• Identify the sample space and events relevant to the problem
• Determine the type of probability distribution that best models the situation (discrete or continuous, specific distribution)
• Use the given information to find the parameters of the distribution (success probability, mean, variance)
• Apply the appropriate probability rules and formulas to calculate the desired probabilities or values
  • Example: using the binomial PMF to find the probability of a specific number of successes in a fixed number of trials
• Utilize conditional probability and Bayes' theorem when dealing with dependent events or updating probabilities based on new information
• Recognize when to use the law of total probability to break down a complex problem into simpler subproblems
• Apply the properties of expectation and variance to solve problems involving random variables
  • Example: using linearity of expectation to find the mean of a sum of random variables
• Interpret the results in the context of the original problem and communicate the findings clearly
• Verify the reasonableness of the solution by checking if the probabilities are within the valid range [0, 1] and if the results make sense intuitively