If you are seeing this message, you may be experiencing temporary network problems. Please wait a few minutes and refresh the page. If the problem persists, you may wish to report it to your local Network Manager.
It is also possible that your web browser is not configured or not able to display style sheets. In this case, although the visual presentation will be degraded, the site should continue to be functional. We recommend using the latest version of Microsoft or Mozilla web browser to help minimise these problems.
Wiley InterScience | ||
 Journal of the Royal Statistical Society: Series D (The Statistician)Volume 46 Issue 1, Pages 1 - 18 Published Online: 5 Jan 2002 2003 Royal Statistical Society
Abstract | Full Text: PDF (Size: 423K) | Related Articles | Citation Tracking  Authorship attribution: the case of Oliver Goldsmith Copyright Royal Statistical Society 1997 KEYWORDS Cluster analysis • Mahalanobis distance • Outlier • Principal component analysis • Selection of variables ABSTRACTIn an author attribution problem where a battery of tests of discrimination is brought to bear on samples of text from a number of different works thought, possibly mistakenly, to be by a given author (in our case, Goldsmith), difficulties arise if the number of samples of text from work known to be that of the given author is small compared with the number of tests. We propose a screening procedure whereby ineffective tests are excluded from subsequent analysis, but the number of tests may still be too large for the application of certain standard types of analysis. We give two methods for dealing with this. Over 100 essays have been attributed to Goldsmith since his death. 12 of these are here submitted to an analysis from which we conclude that several traditional ascriptions are well founded, that three should be considered very doubtful and that there are good grounds for promoting two essays to the canon. |  |
