Читать книгу Applied Regression - Colin Lewis-Beck - Страница 6

Series Editor’s Introduction

Оглавление

Invented more than 200 years ago, apparently independently by the German mathematician Carl Friedrich Gauss and the French mathematician Adrien-Marie Legendre, the method of least squares occupies a central place in statistical methods. Linear least squares regression not only is very widely employed in research but also furnishes a basis for much of applied statistics. Many statistical models — generalized linear models, linear and generalized linear mixed-effects models, survival regression models, and linear structural equation models, to name a few of the more prominent — represent direct generalizations of linear regression; and computation for statistical models often involves least squares fitting — for example, the use of iterated weighted least squares to compute maximum likelihood estimates for generalized linear models. Both for its direct application and for its many generalizations, a sound background in linear least squares regression is fundamental to the study of statistics.

In Applied Regression, Colin and Michael Lewis-Beck provide a thorough primer in linear least squares regression, introducing the method from first principles. They attend to practical details of regression analysis; to the statistical model underlying inference in linear regression and to violations of the assumptions of the model; and — most important – to the interpretation of results and the interplay between statistical modeling and the substance of social research.

There is clearly a need for a brief, accessible, and nontechnical treatment of regression analysis, and the first edition of this monograph was one of the most widely read in the QASS series. I expect that this new, expanded, and extensively revised edition will be similarly well received.

On a personal note, I am particularly pleased to be able to assist in the publication of this monograph because I have known Michael Lewis-Beck since we were both graduate students at the University of Michigan many years ago, and I have subsequently had the pleasure of becoming acquainted with his son, Colin.

John Fox

Series Editor

Applied Regression

Подняться наверх