Gill, Jeff Generalized linear models: a unified approach. (English) Zbl 0963.62060 Sage University Paper Series on Quantitative Applications in the Social Sciences. 134. Thousand Oaks, CA: Sage Publications. 101 p. (2000). This monograph explains and demonstrates a unified approach to applying regression models in the social sciences. Chapter 1 provides an introduction of the generalized linear models and gives an overview of some important concepts plus prerequisites and preliminaries. Chapter 2 introduces the exponential family of distributions. Chapter 3 is devoted to the likelihood theory and the first two moments calculations using exponential families. Chapter 4 is the most important chapter of the monograph, where the generalized linear models are defined and some examples in social sciences given. Chapter 5 develops the statistical computing techniques used to produce maximum likelihood estimates for coefficients in generalized linear models. Chapter 6 discusses the residual analysis and model fit and some examples in social sciences are provided. Chapter 7 concludes the monograph.It is to be noted that the author includes numerous original modeling examples on real world social science data in the monograph. I think that this monograph can serve as a very good reference book for applied statisticians and as a textbook without exercises for those who have a good knowledge of mathematical statistics and linear models at an undergraduate level and want to study generalized linear models. Reviewer: Yuehua Wu (Toronto) Cited in 1 Document MSC: 62J12 Generalized linear models (logistic models) 62P25 Applications of statistics to social sciences 62-01 Introductory exposition (textbooks, tutorial papers, etc.) pertaining to statistics Keywords:exponential families; likelihood; link functions; estimation; Newton-Raphson and root finding; weighted least squares; residuals; goodness of fit; asymptotics PDFBibTeX XMLCite \textit{J. Gill}, Generalized linear models: a unified approach. Thousand Oaks, CA: Sage Publications (2000; Zbl 0963.62060)