Data statistical analysis is essential to research and application in Atmospheric Sciences. Students of this course will learn step by step various theories and methods of modern statistical analysis which usually be applied in atmospheric sciences. Illustrations selected from papers in the journals of atmospheric sciences will be used. Students will be asked to do homework problems set with Matlab Software.
Topics to be covered:

Hypothesis Testing: Background, parametric approaches, nonparametric tests (classical, resampling, the bootstrap, and permutation)

Forecast Verification: Categorical forecasts, probability forecasts, nonprobabilistic forecasts of fields, verification of ensemble forecasts

Multivariate Analysis: Matrix algebra, principal component analysis, canonical correlation analysis, discrimination and classification, cluster analysis

Bayesian Inference: Bayes’ theorem, Bayesian inference with prior distributions, Bayesian prediction (if time permits)

Artificial Neural Network: Concept of ANN, back-propagation ANN, applications

Genetic Algorithm: An introduction (if time permits)

Homework problem sets will be given on a regular basis.

Suggested textbook: Statistical Methods in the atmospheric sciences, D. Wilks, 2nd edition, Academic Press, 2006

Supplementary books: Probability, Statistics, and Decision-making in the Atmospheric Sciences, Chapter 12. A.H. Murphy and R.W. Katz, Eds., Westview Press, 1985

Bayesian data analysis, A. Gelman, et al., Chapman & Hall, 2004