AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Is It a Good Idea to Use a Prior?
There is no way of doing and so what you would typically do is a prior, such as an expedential prior. And then if you say, suppose we know that we're not going to live more than a certain number of centuries ad we'l cut it off there. What we should do is have the prior that went all the way along and diminished and cut that prior and rescale it. But i'm very suspicious over these things with arbitrary intervals where they spread it out uniformly. I think there's an argument that karl shorman made explicitly, that you talked about in that episode,. Where he was saying that a model where an earlier event could crowd out a later event.