Richard F. Gunst

Regression Analysis and Its Application: A Data-Oriented Approach answers the need for researchers and students who would like a better understanding of classical regression analysis. Useful either as a textbook or as a reference source, this book bridges the gap between the purely theoretical coverage of regression analysis and its practical application. The book presents regression analysis in the general context of data analysis. Using a teach-by-example format, it contains ten major data sets along with several smaller ones to illustrate the common characteristics of regression data and properties of statistics that are employed in regression analysis. The book covers model misspecification, residual analysis, multicollinearity, and biased regression estimators. It also focuses on data collection, model assumptions, and the interpretation of parameter estimates.Complete with an extensive bibliography, Regression Analysis and Its Application is suitable for statisticians, graduate and upper-level undergraduate students, and research scientists in biometry, business, ecology, economics, education, engineering, mathematics, physical sciences, psychology, and sociology. In addition, data collection agencies in the government and private sector will benefit from the book.

Praise for the First Edition Statistical Design and Analysis of Experiments "A very useful book for self study and reference." –Journal of Quality Technology "Very well written. It is concise and really packs a lot of material in a valuable reference book." –Technometrics "An informative and well-written book . . . presented in an easy-to-understand style with many illustrative numerical examples taken from engineering and scientific studies." –Choice (American Library Association) Practicing engineers and scientists often have a need to utilize statistical approaches to solving problems in an experimental setting. Yet many have little formal training in statistics. Statistical Design and Analysis of Experiments gives such readers a carefully selected, practical background in the statistical techniques that are most useful to experimenters and data analysts who collect, analyze, and interpret data. The First Edition of this now-classic book garnered praise in the field. Now its authors update and revise their text, incorporating readers’ suggestions as well as a number of new developments. Statistical Design and Analysis of Experiments, Second Edition emphasizes the strategy of experimentation, data analysis, and the interpretation of experimental results, presenting statistics as an integral component of experimentation from the planning stage to the presentation of conclusions. Giving an overview of the conceptual foundations of modern statistical practice, the revised text features discussions of: *The distinctions between populations or processes and samples; parameters and statistics; and mathematical and statistical modeling *The design and analysis of experiments with factorial structures, unbalanced experiments, crossed and nested factors, and random factor effects *Confidence-interval and hypothesis-testing procedures for single-factor and multifactor experiments *Quantitative predictors and factors, including linear regression modeling using least-squares estimators, with diagnostic techniques for assessing model assumptions Ideal for both students and professionals, this focused and cogent reference has proven to be an excellent classroom textbook with numerous examples. It deserves a place among the tools of every engineer and scientist working in an experimental setting.

Regression Analysis and Its Application: A Data-Oriented Approach answers the need for researchers and students who would like a better understanding of classical regression analysis. Useful either as a textbook or as a reference source, this book bridges the gap between the purely theoretical coverage of regression analysis and its practical application. The book presents regression analysis in the general context of data analysis. Using a teach-by-example format, it contains ten major data sets along with several smaller ones to illustrate the common characteristics of regression data and properties of statistics that are employed in regression analysis. The book covers model misspecification, residual analysis, multicollinearity, and biased regression estimators. It also focuses on data collection, model assumptions, and the interpretation of parameter estimates.Complete with an extensive bibliography, Regression Analysis and Its Application is suitable for statisticians, graduate and upper-level undergraduate students, and research scientists in biometry, business, ecology, economics, education, engineering, mathematics, physical sciences, psychology, and sociology. In addition, data collection agencies in the government and private sector will benefit from the book.