Bayesian and Frequentist Regression Methods Website

Bayesian and Frequentist Regression Methods provides a modern account of both Bayesian and frequentist methods of regression analysis. Many texts cover one or the other of the approaches, but this is the most comprehensive combination of Bayesian and frequentist methods that exists in one place. The two philosophical approaches to regression methodology are featured here as complementary techniques, with theory and data analysis providing supplementary components of the discussion. In particular, methods are illustrated using a variety of data sets. The majority of the data sets are drawn from biostatistics but the techniques are generalizable to a wide range of other disciplines. While the philosophy behind each approach is discussed, the book is not ideological in nature and an emphasis is placed on practical application and it is shown that, in many situations, careful application of the respective approaches can lead to broadly similar conclusions. To use this text, the reader requires a basic understanding of calculus and liner algebra, and introductory courses in probability and statistical theory. The book is based on the author's experience teaching a graduate sequence in regression methods.
An introductory chapter describes a number of motivating examples and discusses general issues that need consideration before a regression analysis is carried out. The book is then broken into four parts: I Inferential Approaches; II Independent Data; III Dependent Data; IV Nonparametric Modeling. Exercises reinforce the content of each chapter in the four parts of the book. The first two chapters of Part I provide general descriptions of the frequentist and Bayesian approaches to inference, with a particular emphasis on the rationale of each approach and a delineation of situations in which one or the other approach is preferable. The third chapter in Part I discusses model selection and hypothesis testing. Part II considers independent data and contains three chapters on linear models, general regression models (including generalized linear models) and binary data models. The two chapters of Part III consider dependent data with linear models and general regression models. Mixed models and generalized estimating equations are the approaches to inference that are emphasized. Part IV contains three chapters on nonparametric modeling with a concentration on spline and kernel methods. The examples and simulation studies of the text are almost exclusively carried out using the freely-available R software.


• Table of Contents: here
• Front and Back Cover: here
• The Inevitable Errata: here



Purchasing Information

You can purchase the hard copy or ebook from Springer, or for Amazon go here



R Code for all Analyses and Figures

Chapter 1: Introduction
• R code
  
• Minnesota Lung Cancer Data:   MNlung.txt   MNradon.txt
  
• Spinal Bone Mineral Density Data   spinalbonedata.txt

Part I: Inferential Approaches

Chapter 2: Frequentist Inference
• R code
  
• Minnesota Lung Cancer Data:   MNlung.txt   MNradon.txt
  

Chapter 3: Bayesian Inference
• R code
  
• Minnesota Lung Cancer Data:   MNlung.txt   MNradon.txt
  

Chapter 4: Hypothesis Testing
• R code
  
• Gene expression "pseudo" data (the real data cannot be publicly shared)   GeneSamp1.txt   GeneSamp2.txt
  

Part II: Independent Data

Chapter 5: Linear Models
• R code
  
• Dyestuff Data   dye.txt
  

Chapter 6: General Regression Models
• R code
  
• R code for pharmacokinetic analyses
  
• R code for pharmacokinetic countour plots
  
• Scottish Lip Cancer Data   scotlip.txt
  
• Minnesota Lung Cancer Data:   MNlung.txt   MNradon.txt
  

Chapter 7: Binary Data Models
• R code
  
• BPD and Birthweight Data   birthweight.txt

Part III: Dependent Data

Chapter 8: Linear Models
• R code
  
• WinBUGS code for Dental Example   DentalWinBUGSGirls.txt
• Reduced Dental Dataset   dental-reduced.txt
• WinBUGS code for Dental Example with Reduced Dataset   DentalWinBUGSReduced.txt
• FEV Data   fev.txt The variables are: ID, smoker (1=current, 0=former smoker), times from baseline (years), FEV1.
• WinBUGS code for FEV Example   fevWinBUGS.txt

Chapter 9: General Regression Models
• R code
  
• Contraception Data   contracept.txt The variables are: ID, dose, time, response (experience amenorrhea).
• Seizure Data   seizure.txt
• WinBUGS code for Seizure Example   SeizureWinBUGS.txt
• Minnesota Lung Cancer Data and graph file (for use with INLA)   MNlung.txt   MNradon.txt   MN_county.graph
• Minnesota Spatial Analysis R Code   MNlungSpatial.R
• Minnesota Lung Cancer Shapefiles   MNlung.shx   MNlung.shp   MNlung.prj   MNlung.dbf
• WinBUGS code for Contraception Data Example   ContraceptWinBUGS.txt
• WinBUGS code for Theophylline Data Example   TheophWinBUGS.txt
• Samples to demonstrate flip-flop phenomenon in Fig 9.11   betachain1.txt   betachain2.txt
• Samples from lognormal and power models to produce residual plots Fig 9.13(c)-(f)  TheophWinBUGSLognormalOut.txt   TheophWinBUGSPowerOut.txt
• Samples of parameters from power models to produce plot Fig 9.14  WinBUGSsigma0.txt   WinBUGSsigma1.txt WinBUGSgamma.txt
• Samples of parameters to produce plot Fig 9.15  WinBUGSClmed.txt   WinBUGSkamed.txt   WinBUGSkemed.txt
• Illiteracy data that are analyzed in Exercise 9.5   illiteracy.xls


Part IV: Nonparametric Modeling

Chapter 10: Preliminaries for Nonparametric Modeling
• R code
  
• BPD and Birthweight Data   birthweight.txt

Chapter 11: Spline and Kernel Methods
• R code
  
• Spinal Bone Mineral Density Data   spinalbonedata.txt
• BPD and Birthweight Data   birthweight.txt
• Cosmic Microwave Background Data  cosmicmicrowave.txt

Chapter 12: Nonparametric Regression with Multiple Predictors
• R code
  
• BPD and Birthweight Data   birthweight.txt
• Iris Data   iris.txt