I just love this piece by Chris Sims: "Bayesian Methods in Applied Econometrics, or, Why Econometrics Should Always and Everywhere Be Bayesian", from 2007.
In addition to the solid content, there are some great take-away snippets, such as:
"Bayesian inference is hard in the sense that thinking is hard."
"(People) want to characterize uncertainty about parameter values, given the sample that has actually been observed."
"Good frequentist practice has a Bayesian interpretation."
In addition to the solid content, there are some great take-away snippets, such as:
And Sims' conclusion: "Lose your inhibitions: Put probabilities on parameters without embarrassment."
I can live with that!
© 2013, David E. Giles