ISyE 6416

ISyE 6416: Computational Statistics

Prof. Yao Xie, Georgia Tech, Spring 2015

Welcome to the website for ISyE 6416, Georgia Tech. Computational statistics is an interface between statistics and computer science. This course covers a set of topics including: analysis of simple algorithms such as quick sort and bisection, gradient descent and Newton's method in statistics, linear and logistic regression, discriminant analysis, Gaussian mixture and Hidden Markov models, EM algorithm, Principle component analysis (PCA), model selection and cross validation, bootstrapping, splines, generating random variables, Monte Carlo methods and MCMC.

Course info

  • Instructor: Prof. Yao Xie. Email: yao.xie@isye.gatech.edu. Office Hour: Tues noon at Groseclose 339.

  • Schedule: Tuesday, Thursday, 8:05-9:25am. Room: Instruction Center 213.

  • Syllabus

  • TA: Minkyoung Kang. Email: mkk@gatech.edu. Office Hour: Mon 2-3pm, Friday 2-3pm.

    • Lectures

    1. Introduction

    2. Review of basics: linear algebra, statistics, MATLAB

    3. Analysis of simple algorithms: quicksort and bisection

    4. Newton's method and gradient descent

    5. Fitting linear regression and logistic regression models

    6. Discriminant analysis

    7. Clustering and k-means

    8. Expectation Maximization

    9. Hidden Markov Model

    10. Principle component analysis (PCA)

    11. Model selection

    12. Bootstrapping

    13. Splines

    14. Random number generation

    15. Monte Carlo methods

    Resources

  • Courses
    • Columbia Univeristy
    UBC
    Stanford
  • Books
    • Computational Statistics, James E. Gentle, 2009.
    Pattern recognition, Richard O. Duda, 2000.
    Pattern recognition and machine learning, Christopher M. Bishop, 2007.
    Computational Statistics, 2nd edition, G. Givens and J. A. Hoeting, 2012.
    Machine learning: A probabilistic perspective, K. P. Murphy, 2012.
    The elements of statistical machine learning, 2nd edition, T. Hastie, R. Tibshirani and J. Friedman, 2009.
  • Papers
    • Jackknife, bootstrap and other resampling methods in regression analysis, C. F. J. Wu, Annals Stats, 1986, Vol. 14, No. 4, 1261-1295.
    Mathematics of sparsity (and a few other things), E. Candes, Proceedings of the International Congress of Mathematicians, Seoul, South Korea, 2014.