Intermediate Statistical Methods (Stat 596)
Fall 2017

General Information

Lecturer Zhiqiang Tan
Office: 459 Hill Center
Email: ztan at stat.rutgers.edu

Lectures: MW 10:20-11:40AM, HIL 522
Office hours: M 1:00-1:30PM.
TA Zexi Song
Office: 555 Hill Center
Office hour: W 1:00-1:30PM.
Textbook Venables and Ripley (2002), Modern Applied Statistics with S, Springer.
Hastie, Tibshirani, and Friedman (2013), Elements of Statistical Learning, Springer.
Gelman et al (2014) Bayesian Data Analysis (3rd edition), CRC Press.
Topics
Generalized linear models
Nonparametric regression: smoothing and penalized splines, RKHS
Random and mixed effects, generalized linear mixed models
Computation with latent variables: EM, Markov chain Monte Carlo
Causal inference (if time permits)
Exams and Homework Homework will be assigned. There will be midterm and final projects.
Makeup policy Make-up exams will only be given if written documentation of a major outside circumstance is provided by a dean or a doctor.
Students who miss exams without presenting proper documentation in a timely manner will receive a grade of zero.
Homework Assignments HW 1, HW 2, HW 3, HW 4.
HW 1 files: glm.txt.

HW 1 solution.
HW 2 solution, and a discussion about using a reduced basis matrix and an average response vector with replicated covariate values.
HW 3 solution.
Announcements Sept 18
        Click here for demonstration of various R codes.
Oct 2
        Click here for some R codes using splines. Click here for the LA ozone data.
Oct 18
        Click here for R codes for smoothing splines and RKHS.
Nov 15
        Click here for R codes for linear mixed-effect models, and here for R codes for fitting splines as mixed-effect models.
Dec 7
        Click here and here for R codes for EM, GD, and SA algorithms for fitting Gaussian mixture models.
Dec 11:
        R codes for Gibbs sampling (here) and Metropolis sampling (here) for (toy) simulation from bivariate normal.
        R codes for Gibbs sampling (here) and Metropolis sampling (here) for posterior simulation in the coagulation example (Table 11.2).