SDS 383D, Spring 2017

I am Giorgio Paulon, 1st year PhD student in Statistics at UT Austin. This is my personal page for the Statistical Modeling II class with Prof. James Scott.

Exercise 1

In the first homework, we review some basics of:

• Bayesian updating in simple conjugate families
• Multivariate Normal distribution
• Classical principles of linear regression
• Frequentist approaches to uncertainty quantification

Exercise 2

In the second homework, we study three different models:

• Conjugate Gaussian model with unknown mean and variance
• Conjugate Gaussian linear model
• Heavy tailed linear model (Student t residuals) for outlier detection

Exercise 3

In the third homework, the main topic is nonlinear curve fitting. In particular, the classes of models we analyze are:

• Linear smoothers
• Local polynomial regression
• Gaussian processes for nonparametric regression and spatial smoothing