Philosophical Thoughts

Kevin's philosophical (and probably flawed) thought on Bayesian vs. Frequentist inference:

Frequentist statistics parallel Calvinist and Platonic ideologies. We are either saved or not saved. One is true, predestined to be true by God, although we cannot know the mind of God. Whether you are saved is predestined, but it manifests in the real world as good character, altruism, diligence, piety, wealth, greatness. These are the (fallible) data we can observe in the real world: the man himself, his character, and his deeds. We cannot know for sure that the man is one of the Chosen, but we can guess that there is a better chance that he is chosen than the drunken fool living down the street, based on the data we have. The true parameter (being saved) is fixed and unknowable, and the data, while informative, paint an imperfect picture of the Truth. We can never prove that the man is saved, although we can establish that he is very unlikely to be saved based on the data. Kind of like a null hypothesis. Bayesian inference takes our observations to be true- or as close to truth as we can ever get. Given that truth, we evaluate the utility of various models and hypotheses about the world. There is no unknowable, fixed set of parameters driving the world. There is no God or Platonic Ideal to dictate the Truth with a capital T. We just have our observations to guide our beliefs. Data from our senses, in addition to our preconceived notions of how the world works, are all we have to guide us.

Deep Thoughts, by Jack Handy
"It takes a big man to cry, but it takes an even bigger man to laugh at that man."

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License