Count ordered choices, unordered groups, repeated choices, and arrangements without losing or double-counting outcomes. These tools make finite probability problems much easier to solve.

Apply the rules that every valid probability model must follow, including complements, addition, and inclusion-exclusion. You will learn to spot impossible answers before trusting a calculation.

Calculate probabilities after new information narrows the possible outcomes. Conditional probability becomes the main tool for evidence, diagnosis, filtering, and updating beliefs.

Tell when one event truly gives no information about another, and when that claim is unsafe. You will use independence to simplify problems while avoiding a common source of bad reasoning.

Use Bayes’ rule to reason from effects back to possible causes. The chapter works through false positives, screening tests, spam filters, and base-rate mistakes.

Trace how probability grew from gambling problems into tools for insurance, physics, polling, finance, computing, and risk regulation. The story explains why today’s notation, assumptions, and standards look the way they do.

Represent uncertain quantities as numbers instead of only events. You will create random variables, read their distributions, and connect outcomes to values such as cost, time, count, or score.

Calculate expected value, variance, standard deviation, and long-run payoff. These ideas help compare uncertain choices that have different rewards and risks.

Use Bernoulli, binomial, geometric, negative binomial, hypergeometric, and Poisson models for counts and trials. You will match each model to the kind of real situation it describes.

Work with density curves, cumulative distribution functions, percentiles, and probabilities over intervals. Uniform, exponential, gamma, beta, and related models become tools for measurements, waiting times, and proportions.

Use the normal distribution, standard scores, and normal probability tables or software. You will see why sums of many small effects often look normal and where that shortcut can fail.

Describe several uncertain quantities at the same time with joint, marginal, and conditional distributions. This opens the door to dependence, prediction, and multi-step probability models.

Compute conditional means, conditional variances, and predictions based on partial information. You will use the law of total expectation and the law of total variance to break hard problems into simpler pieces.

Go beyond “positive,” “negative,” or “zero” correlation by using covariance, rank dependence, tail dependence, and copulas. This helps explain why variables can move together in ways a single correlation number misses.

Find the distribution of transformed variables, sums, products, maxima, and added noise. You will use change-of-variable ideas and convolution to build new random quantities from old ones.

Use probability generating functions, moment generating functions, Laplace transforms, and characteristic functions to handle sums and limits. These methods turn many difficult distribution problems into algebra.

Build probability models with priors, likelihoods, posteriors, and predictive distributions. This chapter extends Bayes’ rule from a single calculation into a full modeling framework.

Represent large probability models with diagrams that show which variables directly depend on which others. Bayesian networks, Markov random fields, and factor graphs make complex dependence easier to calculate and check.

Use Markov, Chebyshev, Jensen, Chernoff, Hoeffding, and union bounds when exact probabilities are unavailable. These tools give safe guarantees for risk, error, and worst-case behavior.

Create reproducible simulations with random seeds, sampled distributions, and clear checks against known answers. You will use computation to test intuition and estimate probabilities that are hard to solve by hand.

Estimate probabilities, averages, and integrals with Monte Carlo methods. Importance sampling, stratification, antithetic variables, and control variates show how to get more accuracy from the same computing budget.

Show why sample averages settle down when repeated trials pile up. The weak and strong laws of large numbers explain when observed frequencies can be trusted.

Use the central limit theorem to approximate sums, averages, polling error, measurement error, and aggregate risk. You will also check the conditions that make the approximation reliable.

Work with convergence in probability, almost sure convergence, convergence in distribution, and convergence in mean. These ideas make limit arguments precise and prepare you for advanced probability.

Handle random sequences where order carries no information and symmetry does real work. Exchangeability connects sampling, Bayesian modeling, urn models, and modern distribution-free guarantees.

Use sigma-algebras, probability measures, measurable functions, and integration to remove hidden assumptions from probability arguments. This chapter gives the formal foundation used in advanced research and graduate-level work.

Study what happens when random vectors have many coordinates. Concentration of measure, random projections, and sub-Gaussian behavior explain why high-dimensional systems can be both powerful and surprising.

Model systems that move from state to state with transition probabilities. You will calculate hitting times, stationary distributions, mixing behavior, and long-run averages.

Model random arrivals over time or space, such as calls, defects, particles, claims, or messages. Homogeneous and nonhomogeneous Poisson processes connect counts, waiting times, and independent increments.

Analyze repeated waiting cycles and lines where people, jobs, or packets wait for service. Renewal theory and queueing models help estimate delays, workloads, capacity, and congestion risk.

Work with random processes whose patterns persist over time. Stationarity, autocorrelation, spectral ideas, and simple time-series models help describe signals, demand, weather, markets, and sensor streams.

Use martingales to model fair games, evolving information, stopping rules, and cumulative risk. Optional stopping, martingale convergence, and martingale inequalities become powerful proof and modeling tools.

Connect discrete random walks to Brownian motion and continuous random paths. You will see how scaling limits turn step-by-step randomness into models for particles, prices, noise, and diffusion.

Work with stochastic integrals, Itô’s formula, and stochastic differential equations at a practical first level. These tools support modern models in finance, physics, biology, and engineering.

Estimate the chance of unusually large losses, records, failures, floods, outages, or crashes. Extreme value theory and large deviations focus on the tail events that averages often hide.

Measure uncertainty and information with entropy, cross-entropy, mutual information, and relative entropy. These ideas link probability to compression, communication, inference, and learning systems.

Use randomness to make computation faster, smaller, or more robust. Hashing, Bloom filters, sketches, randomized trials, and concentration bounds show how probability supports real software systems.

Protect individual data by adding random noise with controlled privacy loss. You will connect differential privacy to probability distributions, composition, accuracy tradeoffs, and safe reporting.

Write probability models as programs that can generate data and condition on observations. Modern probabilistic programming connects modeling, simulation, Bayesian inference, diagnostics, and reproducible workflows.

Create prediction sets that keep a promised coverage rate with very few distribution assumptions. Conformal prediction uses exchangeability and calibration data to give practical uncertainty guarantees.

See how modern diffusion models generate data by learning to reverse a gradual noising process. The chapter connects score functions, stochastic processes, and sampling without turning probability into a machine-learning survey.

Follow a full probability project from a messy question to a defensible decision. You will define the target event, choose assumptions, build the model, compute results, test sensitivity, validate against reality, and present the recommendation.

Turn probabilities into clear statements about risk, uncertainty, and limits. This chapter covers intervals, scenarios, visualizations, base rates, rare events, and wording that avoids false confidence.

Map the paths that use probability deeply, including data science, actuarial work, finance, reliability, operations research, epidemiology, AI, engineering, and research. You will plan proof-of-skill projects, useful tools, possible credentials, and habits for staying current.