I often use the term frequentist to describe somebody who uses MLEs for point estimates, confidence intervals for uncertainty in model parameters, and pvalues for hypothesis testing. But have recognized the distinction below and therefore knew the term was inappropriate. Still haven’t found a better term.

From this Gelman discussion about Bayesians vs frequentists:

Bayesian statistics is about making probability statements, frequentist statistics is about evaluating probability statements…Bayesians can feel free, or even obliged, to evaluate the frequency properties of their procedures. Conversely…frequentists can feel free to use Bayesian methods to derive statistical procedures with good frequency properties.

blog comments powered by Disqus


26 June 2013