General Objectives

• Provide an overview of a likelihood framework for developing and evaluating models (as hypotheses) based on the strength of evidence provided by the data, as an alternative to both traditional frequentist statistics and Bayesian methods.
• Give students the knowledge, skill, and confidence to use likelihood methods to enhance their research.

Format and Approach

Daily lectures present principles, while seminars present specific examples from ecological research. These are supplemented by recommended readings from the statistical and ecological literature. See the Course Schedule for details from the most recent course.

Daily lab sessions challenge students to build and parameterize models using likelihood methods. During the first week, lab exercises complement the material presented in morning lectures and seminars. As the course progresses, students spend more of the lab time working on independent projects. The labs are the most important part of the course - our experience is that students only learn the methods through practice. All of the labs and independent projects are done using the R package for statistical computing.

Prerequisites and Intended Audience

The course is intended for graduate students, post-docs, and practicing scientists. An undergraduate or graduate level background in statistics is desired, but the course will teach the basic principles of probability theory required for the methods. Experience with R is useful, but basic skills in R are taught throughout the labs.