Nordic Informatics Network in the Agricultural Sciences

Computer Intensive Statistical Methods - with Biological Applications

How to prepare I (necessary)

Prepare a data set that you would like to use for the course project. The purpose of the course project is to apply either randomisation/bootstrap or MCMC methods (or both) to your own data. Course projects are presented in (informal) lectures at the end of the summer school.

Decide in advance about one or a few problems that you would like to focus on, and make (if necessary) appropriate selections of your full data set. As a general rule it is preferable to have a simpler problem and data set to try out the methods of the course, instead of having a huge and perhaps messy data set where you'll have to spend a lot of time on data cleaning. It may be useful to have analysed the data before with conventional statistical techniques. Make sure to bring the data set in a standard format (e.g., text file, SAS data set, Excel spreadsheet). In case of problems or questions, contact Henrik Stryhn (hes@svs.dk) for guidance on your choice of data set.

How to prepare II (recommended)

• analyse the 2 following simple data sets (to be used in week 1) by standard statistical methods
• Habitat utilization indexes for deciduous woodland of 12 pine martens (in Danish: skovmår):
  0.13, -0.01, -0.01, 0.42, -0.02, 0.01, 0.09, 0.03, 0.04, 0.06, 0.12, 0.03
  The index was constructed so that a value of zero would be expected if an animal used a habitat type in proportion to its occurrence within the animal's territory. The question is therefore what can be said from these data about the pine martens' use of deciduous woodland.
• The percentages of time spent foraging in deciduous woodland by individual pine martens during a radio-tracking study (fictitious data):
  Males: 10, 4, 85, 5, 7, 92, 0, 13.
  Females: 23, 41, 35, 46, 51, 26.
  The problem here is to draw inferences about the relative use of this habitat by the two sexes.
• read chapters 1-2 in the course book for week 2: Markov Chain Monte Carlo in Practice. Eds. Gilks, W.R., Richardson, S. and Spiegelhalter, D., 1996. Chapman and Hall / CRC Press.

How to prepare III (optional)

• download the Minitab 12 demo (to be used in week 1), and get acquainted with the program, for instance using the downloadable tutorials or the Meet Minitab guide.
• download the WinBUGS program (to be used in week 2), and get acquainted with it, for instance using the documentation available on the web-server.

Questions or comments about the content of this page to Henrik Stryhn (hes@svs.dk).

Author: phd@dina.kvl.dk. Updated: 22 June 1999