Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Applied Nonparametric Statistical Methods

Applied Nonparametric Statistical Methods 268 BOOK REVIEWS References Abdrabbo, N. A. and Priestley, M. B. (1969) Filtering non-stationary signals. J. R. Statist. Soc. B, 31, 150-159. Priestley, M. B. (1981) Spectral Analysis and Time Series. London: Academic Press. Priestley, M. B. and Subba Rao, T. (1969) A test for non-stationarity of time series. J. R. Statist. Soc. B, 31, 140-149. J. Pemberton University of Salford Applied Nonparametric Statistical Methods. By P. Sprent. ISBN 0 412 30610 7. Chapman and Hall, London, 1989. x + 260 pp. £14.95. This is a development of an earlier book, Quick Statistics, published in 1981 and the treat­ ment of the material similarly requires a minimum of mathematical expertise. Although the two books cover a large amount of common ground, this one places greater emphasis on the underlying statistics and also reflects the advance in computing practice which has taken place during the last few years. Like the earlier book, this one is well written and easy to read. It begins with a simple introduction to a variety of standard statistical ideas including the notion of nonparametric methods. It then goes on to cover most of the well-known nonpara­ metric methods based on single samples, paired samples, independent samples and more than two samples. The Kruskal-Wallis test, omitted in Quick Statistics, is given adequate coverage here, and a very clear account is given of non parametric methods applied to correlation and regression for bivariate and (very briefly) multivariate data. A section on categorical data includes a short introduction to log-linear models and associated tests, while the two final chapters give an interesting (and illuminating) account of relatively recent developments (robustness, jackknives and bootstraps) and their relevance to this area of statistics, and con­ sider other developments (applications to censored survival data, discrimination, problems involving mixtures of distributions, etc.) not specifically dealt with earlier. A good range of examples is provided for the reader to try, and care has clearly been taken to avoid the usual problems of artificial data. The customary practice of providing solutions to odd-numbered exercises is followed. The illustrative examples used throughout the text have obviously been selected with the intention of providing the greatest possible insight into the procedures and, although a few typographical mistakes were noted, no serious errors of any kind were found. It is unfortunate, though, that a diagram on p. 150 showing Theil regression fitted to a set of data previously fitted by least squares accidentally fails to show the rogue point which is central to the discussion. A strength of the book is that it repeatedly stresses the assumptions which must be satisfied if an application of a particular procedure is to be valid. It provides simple explanations of the rationale behind these procedures, but mostly omits any detailed theoretical treatment, indicating instead where this can be found if needed. An ample bibliography is provided. One surprising omission, perhaps, in a book designed to give a broad introduction to non­ parametric techniques is that of runs tests, which include the best known and easiest to apply tests for randomness in a sequence of observations. But with this one exception, the coverage is good. The book is intended as an introductory text for undergraduates in statistics and as a practical guide to research workers in other disciplines. It fulfils this role excellently and, in addition to providing a good exposition of the standard procedures, Professor Sprent has con­ tributed much useful insight of his own. B.A. Moore Loughborough University a/Technology http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of the Royal Statistical Society Series C (Applied Statistics) Oxford University Press

Applied Nonparametric Statistical Methods

Download PDF
1 page

Loading next page...
 
/lp/oxford-university-press/applied-nonparametric-statistical-methods-jH52oGbPkx

References (9)

Publisher
Oxford University Press
Copyright
© Royal Statistical Society
ISSN
0035-9254
eISSN
1467-9876
DOI
10.2307/2347774
Publisher site
See Article on Publisher Site

Abstract

268 BOOK REVIEWS References Abdrabbo, N. A. and Priestley, M. B. (1969) Filtering non-stationary signals. J. R. Statist. Soc. B, 31, 150-159. Priestley, M. B. (1981) Spectral Analysis and Time Series. London: Academic Press. Priestley, M. B. and Subba Rao, T. (1969) A test for non-stationarity of time series. J. R. Statist. Soc. B, 31, 140-149. J. Pemberton University of Salford Applied Nonparametric Statistical Methods. By P. Sprent. ISBN 0 412 30610 7. Chapman and Hall, London, 1989. x + 260 pp. £14.95. This is a development of an earlier book, Quick Statistics, published in 1981 and the treat­ ment of the material similarly requires a minimum of mathematical expertise. Although the two books cover a large amount of common ground, this one places greater emphasis on the underlying statistics and also reflects the advance in computing practice which has taken place during the last few years. Like the earlier book, this one is well written and easy to read. It begins with a simple introduction to a variety of standard statistical ideas including the notion of nonparametric methods. It then goes on to cover most of the well-known nonpara­ metric methods based on single samples, paired samples, independent samples and more than two samples. The Kruskal-Wallis test, omitted in Quick Statistics, is given adequate coverage here, and a very clear account is given of non parametric methods applied to correlation and regression for bivariate and (very briefly) multivariate data. A section on categorical data includes a short introduction to log-linear models and associated tests, while the two final chapters give an interesting (and illuminating) account of relatively recent developments (robustness, jackknives and bootstraps) and their relevance to this area of statistics, and con­ sider other developments (applications to censored survival data, discrimination, problems involving mixtures of distributions, etc.) not specifically dealt with earlier. A good range of examples is provided for the reader to try, and care has clearly been taken to avoid the usual problems of artificial data. The customary practice of providing solutions to odd-numbered exercises is followed. The illustrative examples used throughout the text have obviously been selected with the intention of providing the greatest possible insight into the procedures and, although a few typographical mistakes were noted, no serious errors of any kind were found. It is unfortunate, though, that a diagram on p. 150 showing Theil regression fitted to a set of data previously fitted by least squares accidentally fails to show the rogue point which is central to the discussion. A strength of the book is that it repeatedly stresses the assumptions which must be satisfied if an application of a particular procedure is to be valid. It provides simple explanations of the rationale behind these procedures, but mostly omits any detailed theoretical treatment, indicating instead where this can be found if needed. An ample bibliography is provided. One surprising omission, perhaps, in a book designed to give a broad introduction to non­ parametric techniques is that of runs tests, which include the best known and easiest to apply tests for randomness in a sequence of observations. But with this one exception, the coverage is good. The book is intended as an introductory text for undergraduates in statistics and as a practical guide to research workers in other disciplines. It fulfils this role excellently and, in addition to providing a good exposition of the standard procedures, Professor Sprent has con­ tributed much useful insight of his own. B.A. Moore Loughborough University a/Technology

Journal

Journal of the Royal Statistical Society Series C (Applied Statistics) – Oxford University Press

Published: Jun 1, 1990

There are no references for this article.