Statistical Methods in the Atmospheric SciencesStatistical Methods in the Atmospheric Sciences, Fourth Edition, continues the tradition of trying to meet the needs of students, researchers and operational practitioners. This updated edition not only includes expanded sections built upon the strengths of the prior edition, but also provides new content where there have been advances in the field, including Bayesian analysis, forecast verification and a new chapter dedicated to ensemble forecasting. - Provides a strong, yet concise, introduction to applied statistics that is specific to atmospheric science - Contains revised and expanded sections on nonparametric tests, test multiplicity and quality uncertainty descriptors - Includes new sections on ANOVA, quantile regression, the lasso and other regularization methods, regression trees, changepoint detection, ensemble forecasting and exponential smoothing |
Contents
| 21 | |
Multivariate Statistics | 551 |
Example Data Sets | 739 |
Probability Tables | 743 |
Symbols and Acronyms | 751 |
Answers to Exercises | 761 |
| 771 | |
| 807 | |
Back Cover | 819 |
Other editions - View all
Common terms and phrases
algorithm American Meteorological Society analysis approach approximately autocorrelation autoregressive average beta distribution binomial bivariate bootstrap calculated Canandaigua climatological clustering coefficient computed confidence intervals correlation corresponding covariance matrix curve data in Table data series data set data values data vectors defined density diagram distance dynamical eigenvalues eigenvectors elements ensemble forecasts ensemble members Equation errors estimated event example exhibit Figure forecasts and observations function gamma distribution Gaussian distribution groups independent indicates Ithaca joint distribution likelihood linear combinations linear regression locations Mahalanobis distance Markov chain maximum methods minimum temperature multiple multivariate null hypothesis pairs parameters plot points posterior distribution postprocessing precipitation predictand predictive distribution principal components prior distribution probability forecasts quantile random relative frequency represent residuals sample mean sampling distribution scalar scatterplot Section skill score spatial squared standard deviation test statistic transformation uncertainty variance variations verification x₁ yields zero