Canadian Strategic Highway Research Program
C-SHRP Bayesian Modelling:
A User's Guide


Table of Contents


PREFACE

ACKNOWLEDGMENTS

EXECUTIVE SUMMARY

CHAPTER ONE - INTRODUCTION

1.1 Purpose
1.2 Scope of the Guide
1.3 Rationale for Using Bayesian Regression in C-SHRP
1.4 Overview of the Bayesian Regression Approach
1.5 Bayesian Regression Software
1.6 Overview of the Bayesian Regression Template

CHAPTER TWO - 10 STEP TEMPLATE - MODEL DESIGN

2.1 Introduction
2.2 Decide What You Want to Model - Step 1
2.3 Select a Dependent Variable - Step 2
2.4 Select Model Type - Step 3
2.5 Select Independent Variables - Step 4
2.5.1 Enumerating Independent Variables
2.5.2 Evaluating Independent Variables Subjectively
2.5.3 Evaluating Independent Variables Using Correlation
2.6 Postulate Functional Form - Step 5

CHAPTER THREE - 10 STEP TEMPLATE - DATA AND ANALYSIS

3.1 Develop Prior & Assemble Data - Step 6
3.2 Procedure for Deriving Data-Based Priors and Model-Based Priors
3.2.1 Data-Based Priors
3.2.2 Model-Based Priors
3.3 Methods for Deriving Subjective Priors
3.3.1 Incremental Orthogonal Method
3.3.2 The Full Matrix Orthogonal Method
3.3.3 The Card Sort Method
3.3.4 The Questionnaire Method
3.4 Procedure for the Full Matrix Orthogonal Method
3.4.1 Developing the Encoding Package
3.4.2 Choose Experts
3.4.3 Conduct Interviews
3.4.4 Validate Results
3.4.5 Analyze Pseudo Data
3.4.6 Assemble Experimental Data
3.5 Perform Bayesian Regression - Step 7
3.5.1 Performing Classical Regression Using XLBayes
3.5.2 Performing an N-Prior Bayesian Regression Using XLBayes
3.5.3 Performing a G-Prior Bayesian Regression Using XLBayes

CHAPTER FOUR - 10 STEP TEMPLATE - MODEL EVALUATION

4.1 Use Model to Predict Performance - Step 8
4.1.1 Compare Model Performance
4.1.2 Selecting a Representative Prior
4.2 Evaluating Model - Step 9
4.2.1 Data/Prior/Posterior PDF Plots
4.2.2 Understanding PDF Plot Results
4.2.3 PDF Plot Results for the Rutting Example
4.2.4 Evaluating the Model
4.2.4.1 Rational Sign
4.2.4.2 Rational Magnitude
4.2.4.3 t-Statistic
4.2.4.4 Which Information Does the Posterior Reflect?
4.2.4.5 Standard Error of the Posterior Model
4.3 Iterate Model - Step 10

APPENDIX A: BAYES THEOREM

APPENDIX B: BAYESIAN REGRESSION THEORY

APPENDIX C : ANSWERS TO COMMON QUESTIONS

Proceed to Chapter One

Return to Main Table of Contents