Data Modeling for the Environmental Sciences
PBIO 295 (3 credits)

Fall 2008 Tuesday / Thursday 2:00-3:15 PM

Instructor: Brian Beckage (Brian.Beckage@uvm.edu)

Location: Discussions & Lectures in Lafayette L100; Labs TBA

Office Hours: By appointment; Marsh Life Sciences 125

Course Description

This seminar course will provide an introduction to data visualization and analysis. This course will emphasize likelihood, information theorteic, and Bayesian approaches to data modeling in an interactive class environment that is based on discussions rather than lectures. This class will also stress applications to real world problems in the environmental sciences. Class time will be divided between discussions (with some lectures) and computer lab. Students will learn the open source R statistical computing language (R project website) and will analyze a data set for a final project. This course is designed to provide students with the analytic tools required for independent research in the environmental sciences.

Course Prerequisites

Intoductory calculus or permission of the instructor.  The course is intended for advanced undergraduates and graduate students.

Course Texts

• We will primarily use a book by Dr. James Clark (Models for Ecological Data) but will also reference a text written by Dr. Michael Lavine (Introduction to Statistical Thought). These texts have not been ordered through the bookstore but can be obtained through the links above or through other book sellers.
• Lab. We will use Dr. Clark's Lab Manual for Models for Ecological Data in lab. (See the link above for Models for Ecological Data.)

Computing

Please install R on the laptop or desktop computer that you use. It would be helpful if you could bring a laptop, with R installed, to our labs. R can be freely downloaded and installed on a wide variety of computing platforms. To obtain R, go to the R project website, follow the link beneath 'Download' on the left, choose a site near your geographic location, download the pre-compiled binary distribution for your operating system, and then follow the instructions for installation. We will also use winBugs software for fitting Bayesian models.

Preliminary Outline of Grading.

Student grades will be based on four components:

1. A final paper that analyzes a data set of the student's choosing. (40% of course grade).
2. A corresponding 20 minute class presentation on this same research topic earlier in the semester (30% of course grade). The presentation should be 15 minutes with 5 minutes for questions. Students should provide an electronic presentation (using powerpoint, keynote, or similar software).
3. Completion of homework assignments will encompass 20% of couse grade.
4. Student participation in class. (10% of course grade) . This will include asking questions and contributing to discussions.
   

Syllabus (tentative):

(C1~Clark Chapter 1; L1~Lab Manual Chapter 1)