Statistical Computing

Overview

This course covers the following topics: (1) Optimization methods including Newton, Fisher-scoring, secant, quasi-Newton, EM, and iteratively reweighted least squares, (2) combinatorial optimization including simulated annealing and genetic algorithm, (3) Simulation and Monte Carlo integration, (4) Markov chain Monte Carlo, and (5) bootstrapping

Assignments

Using Differentials and Generating Data

Iteratively Reweighted Least Squares

EM for Gaussian Mixtures

Sampling Methods: Accept/Reject and SIR

Gibbs Sampling

Exams

Exam 1

Final Exam