The Crisis Of Evidence: Probability & The Nature Of Cause
- 2016-01-15 (Fri.), 10:30 AM
- Recreation Hall, 2F, Institute of Statistical Science
- Prof. William M Briggs
- Dept. of Statistical Science, Cornell University
Abstract
We are living amidst a crisis of evidence, a pandemic of over-certainty, because the nature of probability is generally misunderstood: it is everywhere conditional, not always empirical nor quantifiable, never subjective, separate from decision, and powerless at discerning cause. Yet classical statistical procedures, frequentist or Bayesian, falsely imply they speak about cause, or they claim the impossible and say chance or randomness is a cause. Probability models are only useful at explaining the uncertainty of what we do not know, and should never be used to say what we already know. We should not speak of parameters, which greatly exaggerate certainty; instead, only verifiable predictive methods based on observables should be used.