Bayesian Analysis with Excel and R

Book Description

Leverage the full power of Bayesian analysis for competitive advantage

Bayesian methods can solve problems you can't reliably handle any other way. Building on your existing Excel analytics skills and experience, Microsoft Excel MVP Conrad Carlberg helps you make the most of Excel's Bayesian capabilities and move toward R to do even more.

Step by step, with real-world examples, Carlberg shows you how to use Bayesian analytics to solve a wide array of real problems. Carlberg clarifies terminology that often bewilders analysts, provides downloadable Excel workbooks you can easily adapt to your own needs, and offers sample R code to take advantage of the rethinking package in R and its gateway to Stan.

As you incorporate these Bayesian approaches into your analytical toolbox, you'll build a powerful competitive advantage for your organization---and yourself.

• Explore key ideas and strategies that underlie Bayesian analysis
• Distinguish prior, likelihood, and posterior distributions, and compare algorithms for driving sampling inputs
• Use grid approximation to solve simple univariate problems, and understand its limits as parameters increase
• Perform complex simulations and regressions with quadratic approximation and Richard McElreath's quap function
• Manage text values as if they were numeric
• Learn today's gold-standard Bayesian sampling technique: Markov Chain Monte Carlo (MCMC)
• Use MCMC to optimize execution speed in high-complexity problems
• Discover when frequentist methods fail and Bayesian methods are essential---and when to use both in tandem

Table of contents

Cover Page
About This eBook
Title Page
Contents at a Glance
Copyright Page
Credits
Pearson’s Commitment to Diversity, Equity, and Inclusion
Contents
About the Author
Preface
1. Bayesian Analysis and R: An Overview
    Bayes Comes Back
    About Structuring Priors
    Watching the Jargon
    Priors, Likelihoods, and Posteriors
    Contrasting a Frequentist Analysis with a Bayesian
    Summary
2. Generating Posterior Distributions with the Binomial Distribution
    Understanding the Binomial Distribution
    Understanding Some Related Functions
    Working with R’s Binomial Functions
    Grappling with the Math
    Summary
3. Understanding the Beta Distribution
    Establishing the Beta Distribution in Excel
    Comparing the Beta Distribution with the Binomial Distribution
    Decoding Excel’s Help Documentation for BETA.DIST
    Replicating the Analysis in R
    Summary
4. Grid Approximation and the Beta Distribution
    More on Grid Approximation
    Using the Results of the Beta Function
    Tracking the Shape and Location of the Distribution
    Inventorying the Necessary Functions
    Moving from the Underlying Formulas to the Functions
    Comparing Built-in Functions with Underlying Formulas
    Understanding Conjugate Priors
    Summary
5. Grid Approximation with Multiple Parameters
    Setting the Stage
    Putting the Data Together
    Summary
6. Regression Using Bayesian Methods
    Regression à la Bayes
    Sample Regression Analysis
    Matrix Algebra Methods
    Understanding quap
    Continuing the Code
    A Full Example
    Designing the Multiple Regression
    Arranging a Bayesian Multiple Regression
    Summary
7. Handling Nominal Variables
    Using Dummy Coding
    Supplying Text Labels in Place of Codes
    Comparing Group Means
    Summary
8. MCMC Sampling Methods
    Quick Review of Bayesian Sampling
    A Sample MCMC Analysis
    Summary and Concluding Thoughts
A. Installation Instructions for RStan and the rethinking Package on the Windows Platform
Glossary
Index
Code Snippets

About The Author

Conrad Carlberg