# Teaching

## Introduction to Probability and Statistics: STAT 203

Introduction to discrete and continuous probability distributions, marginal and conditional distributions, distributions of order statistics are taught.

## Linear Models: STAT 256

Introduction to concepts and results in matrix algebra and linear models. Theory and applications of one way anova model, two way anova model, two way anova model with interactions, model estimation and diagnostics, model comparisons are taught.

## Stochastic Process: STAT 243

Introduction to stochastic processes, including Markovian processes, Brownian motion, hidden Markov models and point processes. Details of Gaussian process and its computationally efficient alternatives are also taught. Students are provided with an introduction to Dirichlet process mixture models.

## Advanced Bayesian Computation: STAT 228

Penalized Optimization, Bayesian high dimensional regression, g prior, spike and slab prior, stochastic search variable selection, shrinkage priors sequential monte carlo, assumed density filtering, expectation propagation, stochastic gradient Langevin dynamics, variational inference, stochastic variational inference.

## Gambling and Gaming: STAT 80A

Games of chance and strategy motivated early developments in probability, statistics, and decision theory. Course uses popular games to introduce students to these concepts, which underpin recent scientific developments in eco- nomics, genetics, ecology, and physics. Demonstration and detailed analysis of the game of craps, roulette, poker and blcakjack are presented.

## Introduction to Statistical Learning: STAT 205

This course is designed to teach under graduate and masters students the required knowledge to join industry right after their degree. This course teaches both frequentist and Bayesian methods with data analysis using R and Python. Special emphasis is given on stochastic gradient descent, penalized regression and methods for tackling big data.

## Intermediate Classical Inference: STAT 205B

Statistical inference from a frequentist point of view. Properties of random samples; conver- gence concepts applied to point estimators; principles of statistical inference; obtaining and evaluating point estimators with particular attention to maximum likelihood estimates and their properties; UMVU estimators, minimax estimator; obtaining and evaluating interval estimators; and hypothesis testing methods and their properties.

## Classical and Bayesian Inference: STAT 132/206

This is a calculus-based introduction to statistical inference course. Both, frequentist and Bayesian methods, are presented during the course. The course begins with Bayesian estimation methods: prior, posterior, and different kinds of prior distributions are discussed, define Bayes estimator and what happens in large samples. Then the course moves to maximum likelihood estimators, their properties, and numerical methods. Then the course discusses sampling distributions of statistics: chi-square distribution, sample mean, sample variance, t-distribution, followed by confidence intervals and hypotheses testing. Finally, if time permits, the course discusses simple linear models, study simulation methods and Markov chain Monte Carlo simulation-based inference.