Identification of Outliers by D. M. Hawkins is a monograph on the identification of outliers in statistics. It is edited by D. R. Cox, F. R. S. The book was first published in 1980 by Chapman & Hall and later reprinted by Springer-Science+Business Media. The book discusses the theoretical principles of outlier detection, including measures of performance, optimal tests, and the use of statistics that measure departure from a model. It also covers the performance of outlier tests in normal samples, the gamma distribution, and the detection of multiple outliers. The book also includes non-parametric tests, such as the Mosteller and Doorbos statistics, and discusses the identification of outliers in linear models and multivariate data. The Bayesian approach to outliers is also covered, as well as miscellaneous topics such as discrete distributions and outliers in time series. The book includes appendices with fractiles of various statistical tests for normal samples and other distributions. The author thanks the universities of the Witwatersrand and North Carolina for their support in writing the book, as well as the publishers of several journals for permission to use their material. The book is intended for practitioners interested in applying outlier tests to data and provides a comprehensive overview of the theoretical aspects of outlier detection. The monograph also includes new results and conjectures in the field of outlier theory.Identification of Outliers by D. M. Hawkins is a monograph on the identification of outliers in statistics. It is edited by D. R. Cox, F. R. S. The book was first published in 1980 by Chapman & Hall and later reprinted by Springer-Science+Business Media. The book discusses the theoretical principles of outlier detection, including measures of performance, optimal tests, and the use of statistics that measure departure from a model. It also covers the performance of outlier tests in normal samples, the gamma distribution, and the detection of multiple outliers. The book also includes non-parametric tests, such as the Mosteller and Doorbos statistics, and discusses the identification of outliers in linear models and multivariate data. The Bayesian approach to outliers is also covered, as well as miscellaneous topics such as discrete distributions and outliers in time series. The book includes appendices with fractiles of various statistical tests for normal samples and other distributions. The author thanks the universities of the Witwatersrand and North Carolina for their support in writing the book, as well as the publishers of several journals for permission to use their material. The book is intended for practitioners interested in applying outlier tests to data and provides a comprehensive overview of the theoretical aspects of outlier detection. The monograph also includes new results and conjectures in the field of outlier theory.