Computing In Statistical Science Through Apl

Computing in Statistical Science through APL

A t the terminal seated, the answering tone: pond and temple bell. ODAY as in the past, statistical method is profoundly affected by T resources for numerical calculation and visual display. The main line of development of statistical methodology during the first half of this century was conditioned by, and attuned to, the mechanical desk calculator. Now statisticians may use electronic computers of various kinds in various modes, and the character of statistical science has changed accordingly. Some, but not all, modes of modern computation have a flexibility and immediacy reminiscent of the desk calculator. They preserve the virtues of the desk calculator, while immensely exceeding its scope. Prominent among these is the computer language and conversational computing system known by the initials APL. This book is addressed to statisticians. Its first aim is to interest them in using APL in their work-for statistical analysis of data, for numerical support of theoretical studies, for simulation of random processes. In Part A the language is described and illustrated with short examples of statistical calculations. Part B, presenting some more extended examples of statistical analysis of data, has also the further aim of suggesting the interplay of computing and theory that must surely henceforth be typical of the develop ment of statistical science.

Computing in Statistical Science Through APL

Modern Multidimensional Scaling

Multidimensional scaling (MDS) is a technique for the analysis of similarity or dissimilarity data on a set of objects. Such data may be intercorrelations of test items, ratings of similarity on political candidates, or trade indices for a set of countries. MDS attempts to model such data as distances among points in a geometric space. The main reason for doing this is that one wants a graphical display of the structure of the data, one that is much easier to understand than an array of numbers and, moreover, one that displays the essential information in the data, smoothing out noise. There are numerous varieties of MDS. Some facets for distinguishing among them are the particular type of geometry into which one wants to map the data, the mapping function, the algorithms used to find an optimal data representation, the treatment of statistical error in the models, or the possibility to represent not just one but several similarity matrices at the same time. Other facets relate to the different purposes for which MDS has been used, to various ways of looking at or "interpreting" an MDS representation, or to differences in the data required for the particular models. In this book, we give a fairly comprehensive presentation of MDS. For the reader with applied interests only, the first six chapters of Part I should be sufficient. They explain the basic notions of ordinary MDS, with an emphasis on how MDS can be helpful in answering substantive questions.