In Bayesian analysis we use the Normal-Inverse Wishart distribution for the parameters of multivariate models these prior distributions have some hyperparameters. So how do we find the values of these hyperparameters to get Bayes estimates of the parameters? One way to specify these values as diffuse but for an informative prior how do we find hyperparameters of the prior densities?
Asked
Active
Viewed 637 times
3
-
3Possible duplicate of hierarchical Bayesian models vs. empirical Bayes – Xi'an Jan 13 '17 at 13:38
-
This it the point of doing Bayesian analysis: you use as hyperparameters your best guess at the value and at the uncertainty about your guess. There is no value to "find", just because there is no "true prior". – Xi'an Jan 13 '17 at 17:37