Probability — Advanced
Stochastic Processes and Applications (Graduate Level)
Overview of the Advanced level
The Advanced level covers the theory and applications of stochastic processes—a mathematical framework for handling random phenomena that evolve over time.
Learning goals
- Understand the basic concepts of stochastic processes
- Master the theory of Markov chains
- Understand the construction and properties of Brownian motion
- Understand the representation of stochastic processes via the Karhunen-Loève expansion
- Learn the basics of the Itô integral and stochastic differential equations
- Be able to prove the central limit theorem
- Understand the foundations of information theory
Contents
-
Chapter 1
Stochastic Processes
Definition of a stochastic process, finite-dimensional distributions, sample paths
-
Chapter 2
Markov Processes
Markov property, transition probability matrix, stationary distribution, ergodicity
-
Chapter 3
Brownian Motion
Wiener process, continuity and non-differentiability of paths, reflection principle
-
Chapter 4
Karhunen-Loève Expansion
Optimal orthogonal expansion of stochastic processes, eigenvalue problem of the covariance operator, relation to PCA
-
Chapter 5
Stochastic Differential Equations
Itô integral, Itô's formula, solutions of SDEs
-
Chapter 6
Central Limit Theorem
Characteristic functions, proof of the central limit theorem
-
Chapter 7
Information Theory
Entropy, mutual information, the source coding theorem
-
Chapter 8
Modern Topics
Martingales, the large deviation principle, introduction to random matrices
Martingale Topics
-
Stopping Time
Definition of a stopping time, relation to filtrations, and the bridge to the optional stopping theorem.
-
Optional Stopping Theorem
Conditions for the optional stopping theorem, a proof sketch, and its application to the gambler's ruin problem.
-
Wald's Identity
The expected value of a random sum, $E[S_N]=\mu E[N]$, and its application to sequential analysis.
-
Azuma's Inequality
Concentration of bounded-increment martingales, generalizing the method of bounded differences.
Distribution Reference
-
Stable Distribution
Definition, characteristic function, mean, variance, special cases, and applications of the stable distribution.
-
q-Gaussian Distribution
Definition, PDF, mean, variance, and its connection to Tsallis statistics.
-
Mittag-Leffler Distribution
Definition, PDF, mean, variance, and the role of the Mittag-Leffler function.
-
Tukey Lambda Distribution
Definition, quantile function, mean, variance, and applications.
-
Wigner Semicircle Distribution
Definition, PDF, mean, variance, and its role in random matrix theory.
-
Nakagami Distribution
Definition, PDF, mean, variance, and applications to fading channels.
-
Rice Distribution
The probability density function (with the modified Bessel function $I_0$), parameters, and applications.
-
Burr Distribution
Definition, PDF, CDF, mean, variance, and applications.
-
Dagum Distribution
Definition, PDF, CDF, mean, variance, income distribution, and relation to the Burr distribution.
-
Lomax Distribution
Definition, PDF, CDF, mean, variance, and applications (Pareto Type II).
-
Gompertz Distribution
A survival-time distribution with an exponentially increasing hazard function.
-
Kumaraswamy Distribution
A flexible continuous distribution on the interval $[0,1]$.
-
Irwin-Hall Distribution
Definition, PDF, mean, variance, and connection to the central limit theorem.
-
Bates Distribution
Definition, PDF, mean, variance, and relation to the Irwin-Hall distribution.
-
Delaporte Distribution
Definition, PMF, mean, variance, and applications (a Poisson-negative-binomial mixture).
-
Conway-Maxwell-Poisson Distribution
A generalized Poisson distribution that handles over- and under-dispersion.
-
Beta Prime Distribution
Definition, PDF, mean, variance, and relation to the beta distribution.
-
Exponentially Modified Gaussian Distribution
The convolution of a Gaussian and an exponential distribution.
-
Hyperbolic Secant Distribution
Definition, PDF, mean, variance, and characteristic properties.
-
Logarithmic Distribution
Definition, PMF, mean, variance, and applications (the logarithmic series distribution).
-
Lévy Distribution
Definition, PDF, mean and variance, its property as a stable distribution, and applications.
-
Skellam Distribution
Definition, PMF, mean, variance, and applications (the difference of two independent Poisson variables).
-
Slash Distribution
Definition, PDF, mean, variance, and applications (a standard normal divided by a uniform variable).
-
Variance-Gamma Distribution
Definition, PDF, mean, variance, skewness, and applications.
-
Yule-Simon Distribution
Definition, PMF, mean, variance, and applications.
-
Zeta Distribution
Definition, PMF, mean, variance, and relation to the Riemann zeta function (Zipf's law).
-
Raised Cosine Distribution
Definition, PDF, CDF, mean, and variance.
-
Negative Hypergeometric Distribution
Definition, PMF, mean, and variance.
-
Rademacher Distribution
Definition, PMF, mean, variance, and applications of the $\pm 1$ two-valued distribution.
Reading
-
Continuous Everywhere, Differentiable Nowhere [Reading]
Born from the dance of pollen grains, Brownian motion is a monster that is continuous yet nowhere differentiable. The story of the heart of probability theory, tackled by Einstein, Wiener, and Itô.
-
The Mathematics of a Fair Game [Reading]
In a fair game, no betting strategy can increase your winnings. From the trap of the double-or-nothing strategy to the optional stopping theorem—the story of martingales, which made "there is no clever trick" rigorous.
Prerequisites
- The contents of Probability — Intermediate
- Measure theory and Lebesgue integration
- Basics of functional analysis (helpful)