I have been teaching several undergraduate and graduate courses, including Introduction to Statistics, Introduction to Biostatistics, Advanced Statistical Methods, Generalized Linear Models, Bayesian Analysis, and Statistical Consulting. I have also taught a new elective course on Modern Data Analysis. Playing into my research theme regarding the development of statistical methods for big data problems, this course focuses on machine learning techniques for high-dimensional problems from the statistical point of view. Here are some of the courses I have taught over the past several years.
You can download some of my lecture notes from GitHub.
STATS 200B: Intermediate Probability and Statistical Theory (Mathematical Statistics)
The main focus of this course is parameter estimation for statistical inference. It starts with a review of probability theory. Then, we will discuss some underlying principles of data reduction. Next, we will spend some time on point estimation; we mainly focus on method of moments, maximum likelihood estimation, and Bayes estimation. The last part of the course is devoted to decision theory and different procedures to evaluate estimators. Throughout the course, we will discuss both frequentist and Bayesian perspectives.
STATS 225: Bayesian Analysis
The objective of this course is to explore Bayesian statistical methods and discuss their applications in real life problems. By the end of this course, students would learn how to formulate a scientific question by constructing a Bayesian model, and perform Bayesian statistical inference to answer that question. Although the focus of this course is on Bayesian methodology, throughout this course, students would be also exposed to some theoretical aspects of Bayesian inference. They would also learn several computational techniques commonly used in Bayesian analysis.
STATS 235: Modern Data Analysis (Statistical Machine Learning)
The objective of this course is to introduces statistical methods in machine learning. We will discuss modern statistical methods commonly applied to high dimensional problems, for which traditional methods might fail. By the end of this course, students would learn a range of statistical methods and machine learning techniques, and they would learn how to use these methods to solve complex, high dimensional problems. Throughout this course, students would be exposed to both frequentist and Bayesian paradigms.
STATS 230: Statistical Computing Methods
The objective of this course is to learn computational methods in statistics. By the end of this course, students are expected to be able to write their own computer programs (as opposed to relying on existing packages) to perform statistical analysis. The two main components of this course are optimization methods and sampling algorithms. Primarily, we discuss optimization methods within the frequentist framework and sampling algorithms with respect to their application in Bayesian inference. Additionally, we discuss a variety of other computational methods (e.g., numerical linear algebra, bootstrap) that are commonly used in modern statistics. Students participating in this class are expected to have some background in statistics and probability (at least two quarters of upper division or graduate coursework). Also, they need to be comfortable with at least one programming language.
STATS 211: Statistical Methods II
In this course we cover the following topics: A brief review of linear models; Model assessment and selection; Controlling complexity (PCR, PLS, Lasso); Basis expansions; Splines; GAM; Gaussian process regression models; Multilevel models (mixed effect models); Linear discriminant analysis; Quadratic discriminant analysis; Naive Bayes classifiers.
STATS 212: Generalized Linear Models
In this course we cover the following topics: Linear regression models-- a quick review; A preview of generalized linear models; Exponential family; Binomial, multinomial, and Poisson distributions; Logistic regression for binary outcome; Multinomial logistic models for nominal and ordinal outcome; Poisson regression for counts; Loglinear models; Quasi-likelihood; Bayesian GLM; Random effects models.
STATS 275: Statistical Consulting
The objective of this course is to learn how to form a scientific question, translate it to a statistical problem, apply an appropriate statistical method for inference, and report findings in a language understandable by non-statistician scientists. Throughout this course, students will collaborate with other scientists at UCI and work on a specific project.
STATS 7 & 8: Introduction to Statistics & Biostatistics
This is an introductory level course on statistics covering the following topics: Designing scientific studies; Exploring data; Probability; Continuous and discrete probability distributions; Estimation; Hypothesis testing; Linear regression models. Click here to go to the course website: Stats 8
2222 ISEB, UC Irvine, CA 92697
babaks at uci dot edu