Glim for Ecologists/Book and Disk: Disk included - Rilegato

Crawley, Michael J.

Rilegato

ISBN 10: 0632031565 ISBN 13: 9780632031566

Casa editrice: Blackwell Science Inc, 1993

Riassunto

Learning GLIM will almost certainly change the way you do statistics. The flexibility and generality of the GLIM language encourage a different approach to statistical analysis, which stresses parameter estimation and model criticism rather than hypothesis testing. The GLIM package can handle most of the models the ecologists are likely to fit to their data, including log-linear models of count data using Poisson errors, logistic regression of proportion data using binomial errors, and a wide variety of models for the analysis of survival data. It also allows extremely general approach to more familiar procedures such as analysis of variance, multiple regression and the analysis of contingency tables.
The book provides a practical approach to learning the rationale and methods of linear modelling, starting from absolute basics and progressing in simple steps to cover the critical analysis of experimental and observational data. Minimal background in mathematics or statistical theory is assumed, and the book is aimed at all ecologists from students to experienced researchers. Statistical modelling GLIM is introduced through an extensive series of ecological examples, where graphical display, model criticism, model simplification, plots of residuals and tests of assumptions about error specification are explained in full. The reader is encouraged to work through examples while sitting at a computer, and all of the necessary data, along with numerous exercises and fully worked answers, are provided on a disk supplied at the back of the book.
GLIM for Ecologists is not a statistical recipe book; it informs and educates on the proper use of statistics in ecology, on experimental design and on how to get the most out of your data. Learning GLIM will certainly be worth the effort. It's not the question of finding out how to do the same old tests in yet another new statistics package. On the contrary, it involves mastering a powerful set of tools that were originally developed for professional statisticians, but which are now accessible to ecologists through the simple step-by-step approach adopted in this book.

Su questo libro

Dalla quarta di copertina:

GLIM is one of the most powerful and efficient software packages for statistical analysis commercially avialable. The package can handle most of the analysis that ecologists are likely to carry out: regression, analysis of variance; log-linear models of counts, models in which the variance increases with the mean, models of survival, and much more. One of the objectives of statistical analysis is to distil a long and complicated set of data into a small number of meaningful descriptive statistics. The major drawback with the majority of commercially avilable packages is that they generate huge amounts of data output. GLIM, however, only tells the user what it is explicitly asked for. An initial encounter with GLIM can be a daunting experience for the beginner and a clear and comprehensive introduction to the package is essential. GLIM for Ecologists introduces the reader, step-by-step, to the methods of generalised linear modelling using the GLIM statistical language. Every ecologist intending to use the package will benefit from using the book as a guide through the initially impenetrable GLIM language. The book is appropriate for students and professionals alike and assumes that the reader has a working knowledge of linear regression, analysis of variance, significance tests, running programes on a desktop computer, and directory structure and file management in DOS. The book contains a wealth of exercises and examples and all of the data sets used in the book are available on disc.

Su questa edizione

• Editore: Blackwell Science Inc
• Data di pubblicazione: 1993
• ISBN 10: 0632031565
• ISBN 13: 9780632031566
• Rilegatura: Copertina rigida
• Numero di pagine: 379