Classical and Modern Regression

by Raymond H. Myers

Book Overview

For seniors or graduate students with backgrounds in calculus and linear algebra; concepts are emphasized by using a blend of real data sets and mathematical development.

• Format: Hardcover
• Language: English
• ISBN: 0534921787
• ISBN13: 9780534921781
• Release Date: January 1990
• Publisher: Duxbury Resource Center
• Length: 488 Pages
• Weight: 2.35 lbs.
• Dimensions: 1.0" x 7.7" x 9.5"

Customer Reviews

Excellent Graduate Textbook and Reference

I used this texbook in my graduate Multivariate Analysis class in 2006 and learned a lot from it. It is well organized and addresses typical problems encountered and asked about in multiple linear regression. It has many useful examples, solutions, and exercises. I still use it as a good reference book in an industry company I work for. I highly recommend this book to graduate and postgraduate students.

very nicely presented text on regression by an excellent writer

Myers is a great lecturer and writer. This can be seen in this book, his text on response surface methods and his jointly published books with Doug Montgomery. This is classical linear regression covered in an elementary way for beginning students. Myers is an expert in linear models and response surface methods and it shows in his coverage of these subjects in this text.

Modern Regression Analysis

Linear statistical models for regression, analysis of variance, and experimental designs are widely used today in business administration, economics, engineering and the social, health, and biological sciences. Successful application of these models requires a sound understanding of both the underlying theory and the practical problems that are encountered in using the models in real life situations. While "Classical & Modern Regression with Application" is basically an applied book, it seeks to blend theory and applications effectively, avoiding the extremes of presenting theory in isolation and of giving elements of applications without the needed understanding of the theoretical foundations. In the area of regression analysis, the author have organized the chapters in order to get to multiple linear regression analysis more quickly. The introductory chapters are largely formula based provide principles that have implications for multiple regression. Emphasis is deliberately made on the underlying assumptions of the classical regression model and how to deal with the situations where the data do not abide by these classical assumptions. Model inadequacy, multicollinearity, heteroscedasticity, autocorrelation are some of the problems that are very common in practical data and require specialized knowledge and techniques to cope with. The book presents expanded discussions of regression diagnostics and the corresponding remedial measures in an evolutionary sequence. Starting from the age old techniques of residual analysis, it goes up to the modern robust techniques including robust estimators. In the area of non-linear regression analysis, the author presented models that can be analyzed by adaptation of linear models such as polynomial models, log linear models, dichotomous dependent and independent variables and the "generalized linear model" is used for unbalanced data and the analysis of covariance.On the whole, the book is an excellent addition in the regression analysis literature. It enables the reader to have a clear and vivid picture of the modern regression analysis as how it is evolved and empowers him to analyze the real data with more confidence.