1st Edition

Applied Regression and ANOVA Using SAS



ISBN 9781439869512
Published February 28, 2021 by Chapman and Hall/CRC
384 Pages 100 B/W Illustrations

USD $89.95

Prices & shipping based on shipping country


Preview

Book Description

Designed for researchers primarily interested in what their data are revealing, Applied Regression and ANOVA Using SAS presents rigorous statistical methods without burdening readers with matrix algebra and calculus. Each method is introduced by discussing its characteristic features, the reasons readers would want to use it, and its underlying assumptions. The book then guides readers in applying each method by describing a step-by-step approach, giving SAS code to implement the steps. The SAS code is annotated, which allows users to readily adapt it to their own data set.

In the step-by-step approach, readers are given practical advice on how to evaluate in depth whether the assumptions of a method are reasonable for their data set. The book also gives practical advice on interpreting results in the light of modern multiple testing procedures and simultaneous confidence intervals.

Readers are shown throughout the book how high resolution, publication-ready graphics associated with regression and ANOVA methods are produced with virtually no effort by the SAS user. Suggestions for navigating issues encountered in analyzing real-life data make this book invaluable to both non-statisticians and applied statisticians.

Table of Contents

Review of Some Basic Statistical Ideas. Simple Linear Regression. Additive Multiple Linear Regression. Model Selection. Evaluating a Two-Way Interaction between Continuous Predictors in Multiple Linear Regression. Evaluating a Two-Way Interaction between a Qualitative and Quantitative Predictor in Multiple Linear Regression. Comparing Group Means with One-Way ANOVA. Simultaneous Inference: Methods That Adjust for the Multiplicity Problem. Comparing Group Means Adjusted for Nuisance Variables with Analysis of Covariance. Alternative Approaches When a Linear Regression Model Is Not a Good Fit.

...
View More