Blog Archives

Matlab version of BEST

April 28, 2016
By
Matlab version of BEST

Nils Winter has created a Matlab interface to BEST (Bayesian estimation for two groups of metric-scale data). You can read about it at his GitHub README. 

Read more »

Bayesian estimation of log-normal parameters – Update

April 19, 2016
By
Bayesian estimation of log-normal parameters – Update

Using the log-normal density can be confusing because it's parameterized in terms of the mean and precision of the log-scale data, not the original-scale data. Thus, if your data, y, are nicely described by a log-normal distribution, the estimated m...

Read more »

The remains of R. A. Fisher are visited by Bayesians again

January 29, 2016
By
The remains of R. A. Fisher are visited by Bayesians again

Thanks to Phillip Alday, the second edition of Doing Bayesian Data Analysis has visited the remains of R. A. Fisher at St. Peter's Cathedral in Adelaide. Here is a photo that Phillip snapped: In a previous post, the first edition visited Fisher's ...

Read more »

(A pointer to) bayes.js: A Small Library for Doing MCMC in the Browser

December 31, 2015
By
(A pointer to) bayes.js: A Small Library for Doing MCMC in the Browser

Screen shot from bayes.js blog post.This is just a pointer to a new post at Rasmus Bååth's Blog that might interest readers of DBDA2E. bayes.js: A small JavaScript library that implements an adaptive MCMC sampler and a couple of probability distribut...

Read more »

Lessons from Bayesian disease diagnosis: Don’t over-interpret the Bayes factor, VERSION 2

December 27, 2015
By
Lessons from Bayesian disease diagnosis: Don’t over-interpret the Bayes factor, VERSION 2

[This is a revised version of a previous post which is superseded by this post. This revision has corrected derivations, new R/JAGS code, and new diagrams.] Overview"Captain, the prior probability of thischaracter dying and leaving the showis...

Read more »

Bayesian Item Response Theory in JAGS: A Hierarchical Two Parameter Logistic Model

December 22, 2015
By
Bayesian Item Response Theory in JAGS: A Hierarchical Two Parameter Logistic Model

I recently created a hierarchical two-parameter logistic model for item response theory (IRT). The JAGS script is now in the folder of scripts that accompany the book (available at the book's web site; click book cover at right). Below are slides ...

Read more »

Lessons from Bayesian disease diagnosis: Don’t over-interpret the Bayes factor

December 12, 2015
By
Lessons from Bayesian disease diagnosis: Don’t over-interpret the Bayes factor

This post has been revised and is now superseded by the new post linked >HERE<. Overview"Captain, the prior probability of thischaracter dying is extremely small."A primary example of Bayes' rule is for disease diagnosis (or illicit drug screenin...

Read more »

Prior on df (normality) parameter in t distribution

December 2, 2015
By

Many models in DBDA use a t distribution for "robust" estimation of central tendency parameters. The idea is that the heavy tails of a t distribution can accommodate data points that would be outliers relative to a normal distribution, and therefore th...

Read more »

A case in which metric data are better analyzed by an ordinal model

August 17, 2015
By
A case in which metric data are better analyzed by an ordinal model

Here we consider some data that might have been smoothly distributed over a metric scale, but ended up being concentrated on only a few values. The usual treatment of the data as normally or t-distributed is not appropriate, and instead the data are bi...

Read more »

Bayesian meta-analysis provides high-precision estimate of cosmic impact event at ~12,800 years before 1950

August 3, 2015
By
Bayesian meta-analysis provides high-precision estimate of cosmic impact event at ~12,800 years before 1950

A recently published article in PNAS reports a Bayesian meta-analysis of data from 354 samples from 23 geological sites on 4 continents, yielding a 95% credible interval from 12,835 to 12,735 years before 1950 for the cosmic impact event that formed th...

Read more »


Subscribe

Email:

  Subscribe