LINKS
| Instructor, TAs | Overview | Prerequisites | Class/Lab Times |
| Texts, Datasets | Software | Policies | Grading |
Instructor:
Junjie Zhang, junjie.zhang " at" duke.edu, Ph.D. student, Nicholas School of the Environment and Earth Sciences.
This course will focus on exploratory analysis of data, graphical methods, building and fitting regression models, investigating assumptions, and interpreting results. A major focus of our work will be to learn to write literate data analysis reports, and to develop skills to critique statistical analyses. Most of the examples in the class will come from the environmental sciences.
Topics:
Univariate and multivariate linear regression: assumptions, inference, interpretation, critique, and correction using graphical residual analysis. Analysis of robustness of regression models, use of diagnostics to detect departures from assumptions, and assessment of predictive capabilities of regression models. Model selection: use of information criterion to address the bias-variance tradeoff; Bayesian methods for model selection.
Regression models for other types of response variables, including logistic regression for binary response variables and binary count data and log-linear regression for Poisson counts
Use of a standard statistical software package, S-Plus, for applied data analysis: applications and examples emphasizing the biological and environmental sciences
Prerequisites for STA 242 / ENV 255
D. Moore, G. McCabe. Introduction to the Practice of Statistics. W.H. Freeman & Co. We'll use the 4th edition for this class. Students whose backgrounds in basic statistics are weak should plan to purchase this book and use it to supplement the course textbook.
As prerequisite material, it is assumed that you are familiar with the following material in this book:
The concepts above are covered in the The concepts above are covered in 100-level statistics courses offered in Duke Statistics, as well as STA210.
The School of the Environment gives a statistics diagnostic test to its incoming masters students. It is assumed that you demonstrated proficiency in the topics above on this test.
Class Meeting Times: Tuesdays and Thursdays: 12:40 - 1:55. LSRC A247
Computer Lab Sections: As part of the course, students are expected to attend one of the weekly lab sessions, which will emphasize homework questions and computing issues. There will be 3 sections of lab to be held Wednesdays in 153 LSRC: 8-8:50 am, 9:10-10:00 am, 10:30-11:20 am.
Suggested background text for prerequisite topics:
A course at this level is a prerequisite for this course.
Software: S-Plus Version 6 for Windows, Insightful, Inc.
Obtaining the software -- for Duke students only.
You will be asked to turn in reports using a word processor with an equation editor. If you use Microsoft Word, you can reinstall the software, select "custom" and install the Word Equation Editor add-on. Other equation editor programs such as MathType can be used in conjunction with Word.
Course Resources for Answers to Questions
All written assignments must be done independently. This includes computations, Splus output, graphs, answers to questions and discussion of results.Copying will not be tolerated, and will be treated as a violation of the NSEES honor code. Late work will not be accepted.
Students requesting regrades must make these requests within one week of receiving the graded material. Attach a note explaining the regrade issue to your homework or exam and submit to instructor. The instructor or TA has the option to regrade the entire homework or exam.
Some guidelines for submitting written material:
Last modified: Tue Apr 20 15:17:36 EDT 2004