An Introduction to Bayesian Methods for Ecologists

• Start Date: October 27, 2015
• End Date: October 29, 2015
• Time: 8:30 am- 4:30 pm
• City: Revelstoke, BC
• Venue: Regent Hotel
• Instructor: Dr. Carl Schwarz

Course Description

Bayesian methods supplement the standard analysis methods such as regression, ANOVA, and generalized linear models. They are helpful in two general situations. First, when information from a number of studies is to be merged together. Second, there are certain problems that are “hard” to do using standard methods. For example, dealing with censored data in regression, or random effects in logistic regression. This course will present an overview of the use of Bayesian methods in ecology.

Class size: Class size is limited to 16 people.

Bring: Laptop computer pre-loaded with software (see below)

Pre-Requisites:  Sound understanding of basic regression and ANOVA. For example, you should be able to write down the full statistical model for a linear regression. Have you run a logistic regression before? Some familiarity with R will be helpful as we will use R to access OpenBugs/JAGS after learning the basics.

Course Content

1. Introduction to R
2. Conceptual overview of Bayesian methods and comparisons with standard methods.
3. Introduction to OpenBugs/JAGS
4. Hierarchical models – borrowing information.
5. Two-sample comparisons of means and proportions in a Bayesian context
7. The regression model in a Bayesian context
8. The ANOVA models in a Bayesian context
9. Generalized linear models in a Bayesian context
10. Simple mark-recapture models in a Bayesian context.

Other topics may be introduced depending on the interests of the participants. Please email in advance of the course.

Instructor

Dr. Carl Schwarz, Department of Statistics and Actuarial Science, Simon Fraser University. Carl has taught  many courses with CMI, and is back by popular request!

http://www.stat.sfu.ca/~cschwarz/

Required Software and downloading instructions

Participants should bring a laptop computer pre-loaded with software as noted below.

1. Install R and Rstudio (but not R Commander).
Visit
http://www.stat.sfu.ca/~cschwarz/CourseNotes/HowGetSoftware.html
for installation instructions.

You will find a personal library very convenient (see point 3 in above instructions).

2a. *** WINDOWS *** Install OpenBUGS
Visit
http://www.stat.sfu.ca/~cschwarz/CourseNotes/HowGetSoftware.html
for installation instructions.

Install the following additional packages into R
R2OpenBUGS, coda, digest
Don’t forget to check the box on installing dependencies
when installing the packages.

2b. *** Macintosh *** Install JAGS
Visit
http://www.stat.sfu.ca/~cschwarz/CourseNotes/HowGetSoftware.html
for installation instructions.

Install the following additional packages into R
R2Jags, coda, digest
Don’t forget to check the box on installing dependencies
when installing the packages.

3. Adobe Reader or equivalent to access course material. (Material will be supplied as a download prior to the course).

4. You will need to unzip the course exercises files before the course. They are available here in the Bayesian stats folder.

5. Excel or equivalent for reading data files.

Supplied with course:

• Downloadable manual
• Downloadable sample data sets and course exercises
• Coffee breaks
• Meals and accommodation are not included