Statistics 690STA - Appl Semiparametric Regression

Using data to estimate relationships between predictors and responses is an important task in statistics and data science. When datasets are large, modern methods have been developed that allow us to estimate those relationships without making strong assumptions about those relationships- i.e. we can let the data determine how E(y|x) relates to x. In statistics, these methods are generally referred to as ?nonparametric regression.? This applied graduate course will focus on learning to use nonparametric regression to analyze data. We will read a book, ?Semiparametric Regression with R,? and implement / understand the methods in that book. We will address simple and multiple regression data, binary/count data, spatial data, and correlated/time series data. The course will require a substantial project that will be done in a group of size 3 or more.

STATISTC 625 Note: Prerequisite: Stat 625 or permission of the instructor.
We will use R a lot.
Highly motivated undergraduates who have taken 525 are welcome too.