Dennis Lindley and Tony O’Hagan on YouTube

The great discussion Tony O’Hagan had with Dennis Lindley last March for the Bayes 250 meeting at the RSS is now available on line.

Since this is still close to Dennis’s birthday, I take the opportunity to wish him the best for his 90th birthday.

3 Responses to “Dennis Lindley and Tony O’Hagan on YouTube”

  1. I got to say Lindley doesn’t come off looking as well as I expected in that interview. I take a back seat to no one when it comes to advocating a bayesian understanding of probability. Everyone in this debate should say what they really believe without with caveats or apologies, but refusing to work somewhere because there’s a die hard frequentist working there! What the hell?

    Obviously there was very strong anti-bayesian bias back in his day, but still what the hell? It’s better to have a stand-up debate face-to-face than sniping at each other in obscure journal articles every few years. If I’d been him, I would have wanted to work in departments that only had Frequentists.

    There were some other points too. I get what he was saying about always needing a comparison hypothesis. There’s more truth in what he was getting at then there is in Frequentist cloudlessness about how bayesians can’t really use their equations because they don’t have a exhaustive set of all hypothesis. But he really is kind of wrong about his main point.

    If I produce a physics/statics model in the spirit of Laplace which claims the eccentricity of mars is between .15 and .2 while the data shows it’s between .09 and .095 then I’m going to go ahead and reject the model without having an alternative.

    • While it sounds a reasonable request, I am ambivalent about goodness-of-fit test conducted on their own ground. Having to specify the alternative model (even in a non-parametric format) or the B Plan seems to me like a pertinent requirement.

      • Maybe it just needs a rewording. What I meant was “I reject it in favor or any model that’s not obviously wrong, which will be given at some future date”. However you word it, no one is going to believe a model that explicitly contradicts relevant facts.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.