Fall/Winter 1999-2000:

SCS Seminars and Short Course on Hierarchical/Mixed Models

Reload this page to see the latest changes

Current News

General information

The SCS (Statistical Consulting Service) is holding a short course and seminars on hierarchical/mixed models in the 1999/2000 term. The short course is held on Fridays, November 12, 19 and 26, 1999, from 9 a.m. to 12 noon in Vari Hall 1018. It is necessary to register through the SCS web page. Bi-weekly seminars are held on Fridays from 2:00 to 3:00 starting October 15, 1999. Anyone who is interested is welcome to attend. We intend to work through Bryk and Raudenbush "Hierarchical Linear Models" (see references below). The approach is eclectic, both theoretical and applied. We expect to focus on the applications of hierarchical models from the point of view of a statistical consultant. Please contact Georges.Monette@mathstat.yorku.ca for more information on exact dates and location.

Getting started with S-Plus

The statistical packages used in the course are

S-Plus
SAS
STATA: with the Generalised Linear Latent and Mixed Models (GLLAMM) package

In S-Plus, we use the NLME 3.1 package by Doug Bates and Jose Pinheiro. If you are using S-Plus 4.? under Windows, you need to download and install NLME from the NLME web site. Note that version 2 of NLME is built in to Splus 4.? but version 3.1 has much greater capabilities. When loading NLME in S-Plus be sure to specify first = T:

	> library( NLME, first = T)

so the version 3 functions are used instead those from version 2. The course also uses functions from the MASS library from the book Modern Applied Statistics with S-PLUS by W.N. Venables and B.D. Ripley. You need to download and install the software for windows if you are using S-Plus 4.?. Both NLME 3.0 and MASS are incorporated in S-Plus 2000 and don't need to be downloaded. You might nevertheless want to check the sites above for upgrades.

References

Please send e-mail to:Georges.Monette@mathstat.yorku.ca to add references to this list.

Bryk, Anthony S. and Raudenbush, Stephen W. (1992) Hierarchical Linear Models: Applications and Data Analysis methods. Sage, Newbury Park.
NLME (Non-linear mixed effects) by Bates and Pinheiro and beta release for new version 3.0

Data sets from a course by Nan Laird and David Giltinan
Verbeke, Geert and Molenberghs, Geert (Eds.) (1997) Linear Mixed Models in Practice (A SAS-Oriented Approach). Springer, N.Y.

Littell, R. C., Milliken, G. A., Stroup, W. W., and Wolfinger, R. D. (1996) SAS System for Linear Models. SAS Institute Inc., Cary NC.

Little, R. J. A. (1995) Modelling the drop-out mechanism in repeated measures studies. Journal of the American Statistical Association, 90, 1112-1121.

Pinheiro, Jose C. and Bates, Douglas M. (1998) lme: Linear Mixed-Effects Methods and Classes for S and S-Plus, Version 3.0.

Notes

Week	Topics	Links
Week 1 October 15	Chapters 1 and 2 of Bryk and Raudenbush.
Week 2 October 29
Week 3 November 12
Week 4 November 26
Week 5 December 10
Week 6
Week 7
Week 8
Week 9
Week 10
Week 11
Week 12
Week 13