Statistics in Theory and Practice

Among the topics covered are common probability distributions; sampling and the distribution of sampling statistics; confidence intervals, hypothesis testing, and the theory of tests; estimation (including maximum likelihood); goodness of fit (including c2 and Kolmogorov-Smirnov tests); and non-parametric and rank tests. There are nearly one hundred problems (with answers) designed to bring out points in the text and to cover topics slightly outside the main line of development.

This textbook does a good job of combining the theory behind statistical techniques with practical, worked examples and problems. It's not an introductory statistics book: a working knowledge of basic statistical techniques, calculus and computer programming are needed to take full advantage of the material. It is, however, an excellent textbook/how-to manual for senior undergraduates, graduate students and researchers working in the sciences.

Techniques such as Bayesien statistics and Maximum Likelihood Estimators, the Bootstrap technique, hypothesis testing, and tests of fit are frequently used in the physical sciences. Finding good, textbook style explanations and examples (including worked examples to check your code against) is difficult. This isn't a cookbook in the style of Numerical Recipes, but the examples provided are very useful.

Statistics is a slippery subject, and small errors in understanding can produce meaningless results. This book is a useful tool for making your statisics more accurate in an understandable fashion. I used it extensively during my thesis, and recommend it highly.