Projects - ST 762 - Fall 2009

The assignment: In the chart below, you have been assigned a paper or papers from the recent literature to read. The material in the paper is related to the content of this course. Some of the papers in the list below offer extensions or greater detail on things we have discussed or mentioned in class. Others go beyond the course material but should be understandable by someone familiar with the class material.

You will do two things:

Brief summary paper. Write a brief (no more than 5 double-spaced pages) written summary of your assigned paper(s). The summary should provide, in your own words, a high-level description (no technical details) of what the paper(s) is(are) about and why the work is important, and a brief discussion of how the work is related to/extends topics covered in class. This paper should be something a person with advanced training in statistics and knowledge of the material in this course could go to to get a general idea of what the paper(s) is(are) about and why it is important.
Due date: The day of the presentations for the class, Monday, December 7, 10:00 am - 1:00 pm, in 5270 SAS Hall.
Presentation. You will give a 15 minute oral presentation summarizing the content of your assigned paper(s). Your presentation should have the same goal and be at the same level of detail as your paper.
Due date:: Students will give their presentations to each other on Monday, December 7, 10:00 am - 1:00 pm, in 5270 SAS Hall.

Notes on preparing your paper and presentation: The goal of both the paper and presentation is to present the main, key ideas in your paper(s), so that a reader or audience member will come away with an appreciation for why the paper was written and what the important results are. You will not have space (5 pages) or time (15 minutes) to present all the details and mathematical results, so you should not even try. Thus, for example, you do not need to explain every development and formula in the paper(s); just concentrate on the "big ideas" and the results that are the most interesting, central to the message, and important. Also, you do not need to show proofs of theorems or derivations that may be presented in the paper(s); just give the main results and discuss why they are important. You should aim to give an overall sense of what the paper is about and what its primary message is; a reader or audience member go read the paper if they want to see the details.

As visual aids for your presentation, you should prepare a pdf file or powerpoint file for projection (a pdf file created using LaTeX is strongly preferred). You may wish to visit the ST 810A web page for information on using LaTeX to make slides with seminar.sty and tips on giving oral presentations.

Your grade for the paper and presentation will depend on how well you achieve these objectives and on the quality of delivery of your presentation.

Important: Throughout the semester, we have developed certain notation. Thus, your paper and presentation will be most effective for this particular audience if you present the material in your assigned paper(s) in the notation we have used in class. The notation used in your assigned paper(s) may very well be different from that we have used. Wherever quantities in the paper(s) are the same as quantities we have discussed, you should use the notation we have developed in class.

Because journal space is limited, it is common for details to be left out and reference made to previously-published papers where the details may be found. The paper(s) to which you have been assigned may reference other papers. If this is the case, it is your responsibility to seek out these references and determine whether what they say is of importance in presenting the main ideas in your assigned paper.

Please see me if you have any questions about your paper(s) or want feedback on what to present.

Student Paper(s)

David Vock Heagerty (1999), Marginally specified logistic-normal models for longitudinal binary data, Biometrics, and Heagerty and Zeger (2000), Marginalized multilevel models and likelihood inference, Statistical Science

Melinda Thielbar Zeng and Davidian (1997), Bootstrap-adjusted calibration confidence intervals for immunoassay, JASA

Ian Fiske Kauermann, G. and Carroll, R.J. (2002), A note on the efficiency of sandwich covariance estimation, JASA

Zifang Guo Pepe and Anderson (1994), A cautionary note on inference for marginal regression models with longitudinal data and general correlated response data, Communincations in Statistics, and Pan, Louis, and Connett (2000), A note on marginal linear regression with correlated response data, American Statistician

Justin Replogle Mancl and DeRouen (2001), A covariance estimator for GEE with improved small-sample properties, Biometrics, and Pan (2001), On the robust variance estimator in generalized estimating equations, Biometrika.

Phil Schulte Vonesh, Wang, Nie, and Majumdar (2002) Conditional second-order generalized estimating equations for generalized linear and nonlinear mixed effects models, JASA

Eric Reyes Lipsitz, Stuart R., Laird, Nan M., and Harrington, David P (1991) Genearlized estimating equations for correlated binary data: Using the odds ratio as a measure of association, Biometrika, and Carey, Vincent, Zeger, Scott L., and Diggle, Peter (1993) Modeling multivariate binary data with alternating logistic regressions. Biometrika.

Yu Zhang Wang, L. and Zhou, X-H. (2007) Assessing the adequacy of variance function in heteroscedastic regression models. Biometrics.

Student	Paper(s)
David Vock	Heagerty (1999), Marginally specified logistic-normal models for longitudinal binary data, Biometrics, and Heagerty and Zeger (2000), Marginalized multilevel models and likelihood inference, Statistical Science
Melinda Thielbar	Zeng and Davidian (1997), Bootstrap-adjusted calibration confidence intervals for immunoassay, JASA
Ian Fiske	Kauermann, G. and Carroll, R.J. (2002), A note on the efficiency of sandwich covariance estimation, JASA
Zifang Guo	Pepe and Anderson (1994), A cautionary note on inference for marginal regression models with longitudinal data and general correlated response data, Communincations in Statistics, and Pan, Louis, and Connett (2000), A note on marginal linear regression with correlated response data, American Statistician
Justin Replogle	Mancl and DeRouen (2001), A covariance estimator for GEE with improved small-sample properties, Biometrics, and Pan (2001), On the robust variance estimator in generalized estimating equations, Biometrika.
Phil Schulte	Vonesh, Wang, Nie, and Majumdar (2002) Conditional second-order generalized estimating equations for generalized linear and nonlinear mixed effects models, JASA
Eric Reyes	Lipsitz, Stuart R., Laird, Nan M., and Harrington, David P (1991) Genearlized estimating equations for correlated binary data: Using the odds ratio as a measure of association, Biometrika, and Carey, Vincent, Zeger, Scott L., and Diggle, Peter (1993) Modeling multivariate binary data with alternating logistic regressions. Biometrika.
Yu Zhang	Wang, L. and Zhou, X-H. (2007) Assessing the adequacy of variance function in heteroscedastic regression models. Biometrics.