Skip to main content
main-content

Über dieses Buch

The new version has two additions. First, at the suggestion of Stephen Stigler I we have replaced the Table of Contents by what he calls an Analytic Table of Contents. Following the title of each section or subsection is a description of the content of the section. This material helps the reader in several ways, for example: by giving a synopsis of the book, by explaining where the various data tables are and what they deal with, by telling what theory is described where. We did several distinct full studies for the Federalist papers as well as many minor side studies. Some or all may offer information both to the applied and the theoretical reader. We therefore try to give in this Contents more than the few cryptic words in a section heading to ~peed readers in finding what they want. Seconq, we have prepared an extra chapter dealing with authorship work published from. about 1969 to 1983. Although a chapter cannot compre­ hensively Gover a field where many books now appear, it can mention most ofthe book-length works and the main thread of authorship' studies published in English. We founq biblical authorship studies so extensive and com­ plicated that we thought it worthwhile to indicate some papers that would bring out the controversies that are taking place. We hope we have given the flavor of developments over the 15 years mentioned. We have also corrected a few typographical errors.

Inhaltsverzeichnis

Frontmatter

Chapter 1. The Federalist Papers As a Case Study

Abstract
When two statisticians, both flanks unguarded, blunder into an historical and literary controversy, merciless slaughter is imminent. Our persistence needs explanation.
Frederick Mosteller, David L. Wallace

Chapter 2. Words and Their Distributions

Abstract
When we leave general style as a poor bet and pay attention to words, we find that Hamilton and Madison use certain words at quite different rates. Douglass Adair brought this spectacularly to our attention by pointing out their uses of while and whilst. In our work, we have used individual words as the principal basis for measuring likelihood of authorship. Early investigations convinced us that most single variables, carefully selected or not, have little discriminating value, and that a large pool of variables provides the greatest hope of success. Sentence length is a good example of a stylistic variable which had even been used effectively elsewhere, yet failed miserably here. Since the rate for each word can be regarded as a variable, words supply a pool of thousands of variables. Furthermore, words are easily recognized and effective for discrimination.
Frederick Mosteller, David L. Wallace

Chapter 3. The Main Study

Abstract
In this chapter, we present the methods and results of the main study. To simplify the exposition, we describe the methods only for the simpler model based on Poisson distributions of word frequencies but we give the numerical results for both the Poisson and the negative binomial. The technical development for the full model based on negative binomial distributions is postponed to Chapter 4, along with detailed mathematical treatments of special problems.
Frederick Mosteller, David L. Wallace

Chapter 4. Theoretical Basis of the Main Study

Abstract
This chapter is a sequence of technical appendices to Chapter 3. These appendices indicate solutions to some methodological problems that arise in applying Bayes’ theorem to the analysis of data. In a textbook, Chapter 4 would be starred; the chapters that follow are not dependent on the material in this chapter, and most readers would do well to skip it on first reading. Unlike the other chapters, this chapter is written only for those seriously interested in statistical theory. We hope that, beyond documentation, the solutions presented may stimulate others to make improvements, generalizations, or to create analogous solutions for new problems.
Frederick Mosteller, David L. Wallace

Chapter 5. Weight-Rate Analysis

Abstract
While our main study is Bayesian in character, we want also to see how a more traditional approach handles problems of discrimination. Some readers may prefer such an analysis, and many will wish to compare the results achieved by the two methods.
Frederick Mosteller, David L. Wallace

Chapter 6. A Robust Hand-Calculated Bayesian Analysis

Abstract
Robustness is a term attached to methods that are insensitive to assumptions extraneous to what is being studied. For example, in continuous distributions the shape of the distribution of sample medians is insensitive to the shape of the tails of the distribution from which the sample is drawn, and therefore medians are robust against changes in tails. Our main study depends upon distributional assumptions, such as the Poisson or negative binomial, and though we have studied their appropriateness, still it would be well to have a method that is less sensitive to distributional shape. Of course, we cannot expect from the robust study the strength of discrimination of the main study.
Frederick Mosteller, David L. Wallace

Chapter 7. Three-Category Analysis

Abstract
For a long time, we could not find an economical way to carry out a robust Bayesian analysis, but we did see a fairly straightforward way to do a classical analysis, using categories rather than measured rates. The latter study was executed before the robust Bayesian analysis was begun, and so we report it here, although in many ways the two studies are similar.
Frederick Mosteller, David L. Wallace

Chapter 8. Other Studies

Abstract
While working on the general discrimination problem, we did some studies that do not fall in the main line of research reported in this volume. Some of these studies are reported here. Section 8.2 suggests a method for getting started on authorship problems; the method draws on our experience but does not employ all our paraphernalia.
Frederick Mosteller, David L. Wallace

Chapter 9. Summary of Results and Conclusions

Abstract
Our remarks fall under four heads: The Federalist study, authorship problems generally, discrimination and classification problems, and Bayesian studies.
Frederick Mosteller, David L. Wallace

Chapter 10. The State of Statistical Authorship Studies in 1984

Abstract
By adding this chapter on authorship problems 20 years after the original appearance of this book, we give a general idea of the state of the field, of its strengths and weaknesses, and of where some challenging problems and useful work on authorship might lie. Because whole books are now appearing in the area, we want only to portray the state and thrust of the art of authorship resolution. Although we may make an occasional critical remark, our intention is to contribute to the overall assessment, rather than to a specific study. Essentially we treat the time period since Bailey’s (1969) review paper.
Frederick Mosteller, David L. Wallace

Backmatter

Weitere Informationen