Stat 230a: Linear Models
UC Berkeley
Offerings
Overview
Theory of least squares estimation, interval estimation, and tests under the general linear fixed effects model with normally distributed errors. Large sample theory for non-normal linear models. Two and higher way layouts, residual analysis. Effects of departures from the underlying assumptions. Robust alternatives to least squares.
Logistics
Three hours of Lecture and Two hours of Laboratory per week for 15 weeks.
Prerequisites
Matrix algebra, a year of calculus, two semesters of upper division or graduate probability and statistics.