Estimation of functional parameters appears in many applications and statistical problems from probability density estimation to regression, time series analysis, and econometrics. This course covers some classical techniques in the area such as kernel and orthonormal basis estimation as well as more modern methods like neural networks. Model selection and parameter tuning techniques including cross-validation and penalization are also addressed. Prerequisites: Math 3200 and Math 493, or Math 3211, a good course in linear algebra (Math 309 or 429), and some fundamentals of computer programming (CSE131).
Course Attributes: FA NSM; AS NSM