Statistics 604: Introduction to Computational Statistics

Fall 1999 Syllabus, Prof. H. Joseph Newton

Content of the Course

  1. Essential algorithms used in carrying out statistical methods, including:

    1. Evaluating pdf, cdf, and quantile functions of distributions.
    2. Interpolating, smoothing, and numerically integrating functions.
    3. Random number generation and Monte Carlo simulation.
    4. Sorting, searching, and data base manipulations.
    5. Matrix algebra and its application to regression and analysis of variance.
    6. Nonlinear optimization.
    7. Algorithms in mutivariate analysis.
    8. Algorithms in time series analysis.
    9. Computer graphics.

  2. Creating statistical software on graphics workstations (PC's and SUN's), including:

    1. Introduction to Fortran (and maybe C).
    2. Writing simple interactive graphics programs.

  3. Using "matrix oriented" software, including Splus.

  4. Using the TeX text formatting and typesetting program.

  5. Becoming familiar with our Unix workstation network including e-mail, ftp, statlib, etc.

Course Materials

I have prepared a set of notes to be used in the course. They are on the STAT 604 web page and can be printed, either by chapter or all together.

Prerequisites

Knowledge of distribution theory and regression analysis (using matrices) such as covered in STAT 601 is assumed. Previous exposure to a high level programming language such as C, Pascal, or Fortran is helpful but not required. No previous exposure to matrix oriented languages is assumed.

Determining the Course Grade

There will be two in-class, closed-book exams and a final. Each of these three exams will be worth 20% of the grade and will cover algorithms primarily. There will also be a series of programming projects, worth a total of 40% of the grade.