ENVS1134 Statistical methods for analyzing environmental data (5 cr)

An advanced course related to analyzing environmental and spatial data with R Statistics software. Main topics include statistical inference, multivariate methods (for modeling, classification, clustering, and dimension reduction), meta-analysis, analyses of time series, community composition analyses, and analysis methods for spatial data.

After completing this course, the student should be able to

• Study the assumptions of statistical methods, like normality and homoscedasticity.
• Study bivariate correlations using relevant methods for different data types.
• Fit regression models to data containing independent observations, and perform variable and model selection.
• Fit non-linear dose – response models.
• Use regular statistical tests, like analysis of variance and non-parametric methods, correctly, and perform multiple comparisons.
• Analyze data obtained from repeated measurements studies.
• Calculate effect sizes and response ratios, and combine results from several research studies with suitable meta-analysis methods.
• Describe the basic principles of statistical learning, and resampling methods.
• Use classification, clustering, and projection methods for high-dimensional data.
• Use methods related to the analysis of time series, both in time domain and in frequency domain.
• Use community composition analysis methods.
• Read spatial data into R, and perform some rather simple spatial analysis methods with it.
• Use the R Statistics software well enough to solve an exercise work with it.

TILP2500, TILP2600 (Basics of statistics), and BENA4033 Basics of geoinformatics (or other basic knowledge of geoinformatics).

Lectures and exercises (including their solutions) will be put into Moodle.