NUMERICAL ANALYSIS OF ENVIRONMENTAL DATA
1st March - 12th March 1999 (Professor H.J.B. Birks & Dr. M. Kernan)
Course Fees: 500.00 UK pounds
Biological and environmental data are usually complex, consisting of many observations and many variables. This course provides an overview of the main techniques of multivariate data analysis that are relevant and useful in ecology and in the study of environmental change, particularly using ^Sproxy^T biostratigraphical data. Emphasis is on statistically robust and ecologically realistic numerical techniques for both descriptive and hypothesis-testing purposes. The course covers Exploratory data analysis, Classification, Regression analysis, Indirect gradient analysis (ordination), Direct gradient analysis, Calibration and transfer functions, Analysis of temporal and spatial data, Discriminant analysis and Hypothesis testing.
The course assumes that the course participants have some familiarity with basic statistical concepts and methods and with the use of personal computers. You should be familar with concepts such as least-squares linear regression, correlation, simple analysis of variance, sum of squares, chi-square tests, and basic graphics.
For more information and registration forms, please contact:
Catherine Dalton
E-mail: c.dalton@ucl.ac.uk
Tel: +44 (0)171 380 7575
Fax: +44 (0)171 380 7565
Or view our web site:
http://www.geog.ucl.ac.uk/ecrc/teaching.htm