Instructor: Harrison H. Zhou
Class Time: Thursday 4:00-6:00PM in Room 107, 24 Hillhouse Ave (Statistics department building)
Course Description: Introduction to nonparametric methods such as kernel estimation, Fourier basis estimation, wavelet estimation. Optimal minimax convergence rates and constants for function spaces, with connections to information theory. Adaptive estimators (e.g., adaptive shrinkage estimation). If time permits: high dimensional function estimation, functional data estimation, classification, or nonparametric asymptotic equivalence. Applications to real data. Some knowledge of statistical theory at the level of STAT 610a is assumed.
"Function estimation and Gaussian sequence model" by Iain Johnstone.
Introduction to Nonparametric Estimation by Alexandre Tsybakov (online version available)
"All of nonparametric statistics", by Larry Wasserman.
To pass this course, you need to give a presentation and attend at least 10 classes.
Draft Lectures (Many mistakes are expected!):
Go to classes*v2. All draft lectures are in Resources. Many mistakes are expected!