Bayesian Analysis (Inverse Probability)

Page 1

"If you are not using an Informative Prior, you are leaving money on the table" -Robert Weiss

Why use Bayesian data analysis?

You have great flexibility when building a models, and can focus on that, rather than computational issues.

You can include information sources in addition to the data, for example, expert opinion.

The result of a Bayesian analysis retains the uncertainty of the estimated parameters, which is very useful in prediction and decision analysis

Why not use Bayesian data analysis?

Everything is just working fine as it is.

You are not that interested in uncertainty.

It's too computationally demanding.

Too much work setting up.