Welcome to 36-463/663, Hierarchical and Multilevel Models!

Here are a few things you should know about our course before the
semester begins.  

***NOTE*** AT THE END OF THIS EMAIL IS A PRE-CLASS ASSIGNMENT THAT
           YOU NEED TO COMPLETE BY 9AM TUESDAY!

* WEB PAGE

The main web page for this course will be

   http://www.stat.cmu.edu/~brian/463

There is also a blackboard page, for turning in assignments, bulletin
board, and reporting grades.

* PREREQUISITES

  You must have already taken, or be currently enrolled in, 36-401
  (linear regression) or a similar class.

  I will assume I can talk to you about statistics at the level of
  36-226 (calculus-based statistics) and 21-240 or 21-241 (matrix
  algebra).

  You will also be expected to know, or learn quickly, the
  computational software used for this course.  That is, primarily R
  and WinBUGS.

  Most assignments will be turned in electronically.  pdf is the
  preferred file format.  

   - IF YOU KNOW MSWORD: it is a good idea to know how to convert word
     files to pdf.

   - IF YOU KNOW LaTeX: pdflatex or a similar command produces latex
     automatically from LaTeX source.

* TEXTBOOKS

  The required text for this course is

  * Gelman, A. & Hill, J. (2007). Data Analysis Using Regression and
    Multilevel/Hierarchical Models.  NY: Cambridge Univ Press.

  I recommend that you also find a copy of 

  * Lynch, Scott M. (2007). Introduction to Applied Bayesian Statistics
     and Estimation for Social Scientists. New York: Springer.
  
  A pre-publication draft used to be availabe at
  http://www.princeton.edu/~slynch/bayesbook.pdf, but it seems to have
  disappeared.  I will try to make relevant materials available as a
  handout if you do not want to buy a copy for yourself.

  Most other materials for the course will be available on-line.

* COMPUTING

 I anticipate that this course will depend strongly on three software
 packages:

 (1) R, a statistical analysis and programming environment.       
     See http://cran.r-project.org/

 (2) WinBUGS, a package for doing Bayesian statistics.       
     See http://www.mrc-bsu.cam.ac.uk/bugs/

 (3) Rube, for using WinBUGS from R
     See http://www.stat.cmu.edu/~hseltman/rube/

 I strongly prefer you to run all of this in Windows rather than the
 Mac OS or Linux, because the three packages above (and especially
 WinBUGS) work together best in Windows.

 If you have already installed R on your laptop, make sure to upgrade
 it to the current version (2.15.1).  See the following page for details:
 http://www.stat.cmu.edu/~brian/463/install-R.html

 If you use a Mac, you will have to learn to run Windows on your Mac,
 and install R and WinBUGS in Windows.  Some instructions/suggestions
 are available at http://www.stat.cmu.edu/~brian/463/winmac.html.

 You will very soon need to install WinBUGS and rube

 (see http://www.stat.cmu.edu/~brian/463/install-winbugs.html   
 and http://www.stat.cmu.edu/~brian/463/install-rube.html)

 However, this does not have to be done in the first week of classes.

* PRE-CLASS ASSIGNMENT

 Please complete this assignment by Tues Aug 28, 9AM.

 1. Install or update your version of R to the current version
    (2.11.1).  The instructions/experience below may be helpful:
        http://www.stat.cmu.edu/~brian/463/install-R.html

 2. Open R, by clicking on its icon on your desktop, and open
    http://www.cs.cmu.edu/~10702/R2/Rintro.pdf so you can read it.

    * If you have not used R very much in the past, type in all of
      the examples in Rintro.pdf

    * If you already feel comfortable in R, read the examples in
      Rintro.pdf, and try to predict what will happen.  Type in any
      examples where you are not sure.

 3. The "turn in" part.  Go to the 36-463 Blackboard site (shows up as
    "F12-Hierarchicl and Multilevel Models"), go to Discussion Board.
    Under the "Pre-Class Assignment" Forum, each student should start
    a new thread with a message containing the following THREE pieces
    of information:

    * The class or project in the past where you have used R the
      most.  If you have never (or hardly) used R before, *please*
      tell me that too!

    * Any questions or surprises you run into, as you go through the
      Rintro.pdf examples above.

    * A description of your last encounter with linear regression
      and/or mixed effects linear models.

    This has to be done by 9am Tues Aug 28, so that I have time to
    look through it before class!

Looking forward to seeing you in class!

-BJ

Brian Junker                    (412) 268 - 2718
Department of Statistics        brian@stat.cmu.edu
232 Baker Hall                  FAX: (412) CMU-STAT
Carnegie Mellon University        or (412) 268-7828
Pittsburgh PA 15213 USA

WWW: http://www.stat.cmu.edu/~brian/