Statistical Optimization

1 credits · Undergraduate

The principles and practice of statistical optimization for modern data analysis. Linear algebra for statistical modeling. Numerical optimization, including gradient-based methods and Newton-type algorithms. Expectation-Maximization and iterative estimation for latent variable models. Matrix decompositions and regularization. Simulation and algorithmic implementation in computational statistics. Prerequisites: STA 221L or STA 210L and (MATH 216, 218D-1, 218D-2, or 221).