Description
This course offers an introduction to probability models and simulation methods, emphasizing the integration of theoretical foundations with practical applications. Topics include generative classification models using the multivariate normal distribution, binomial mixture models, Poisson processes, and Markov chains. In addition to model development, the course introduces techniques for generating random samples and employing them in simulation studies.
Emphasis is also placed on the application of these models to real-world datasets using the R programming language.
Prerequisites
1 course with a minimum grade of C- from (Stat410); or 1 course with a minimum grade of C- from (Stat400).
Level of Rigor
Advanced
Sample Textbooks
A First Course in Probability, by Sheldon Ross, Stochastic Processes, Sheldon Ross, Statistical Inference, Casella& Berger, additional class notes.
Applications
The topics covered in this course collectively provide a foundation for modeling and analyzing complex data through probabilistic methods and simulation. These models are useful in practical settings that involve classification, clustering, and decision-making by leveraging distributions, mixture models, and other generative approaches. Random sampling techniques allow for simulations and computational inference, which are central to modern data science and statistical practice. Poisson processes and Markov chains help model dynamic systems over time, with applications in fields such as finance, engineering, biology, and operations research.
If you like this course, you might also consider the following courses
Stat420, Stat426
Additional Notes
Students interested in grad school in AMSC should consider this course
Students interested in grad school in MATH should strongly consider this course
Students interested in grad school in STAT should strongly consider this course
Topics
Probability Background:
- Common Probability Distributions
- Conditional Expectation and Conditional Variance
- Mixture Distributions
- Bivariate and Multivariate Normal Distribution
Generative Classification Models
- Naive Bayes
- Linear and Gaussian Discriminant Analysis
- Understanding decision boundaries
Mixture Models
- Poisson Binomial Mixtures, some variations and their properties
- Beta-Binomial Mixtures, some variations and their properties
Generating Random Samples
- Generating random samples using the inverse transform method.
- Accept-Reject sampling strategy.
- Methods to generate normal random samples.
- Implementing samplers using R.
Poisson Processes
- Poisson and Exponential distributions.
- Equivalent definitions for Poisson processes.
- Properties of Independent Poisson processes
- Branching processes
- Conditional distributions on arrivals and wait times.
- Simulations and applications of Poission processes.
Markov Chains
- Markov Property
- Kolmogorov-Chapman equations
- Classification of states
- Limiting probabilities
- Mean time spent in transient states
- Branching processes
- Applications and simulations.