- Cardinal or measurable probability. This would be the domain of risk proper. The easiest would be the probability of a dice. This could also be considered actuarial risk. I would also classify this as any risks that are countable. In actual reasoning, exact measures of this kind will occur comparatively seldom. My this reasoning risk management would be very difficult because there are many risks which are just not countable.
- Ordinal probability or relative position of the event in a ranking. Something is more likely to occur but not how much more likely. This is the domain of uncertainty or vague knowledge. Most of the risks that we face would fall into this category. It is more likely to rain today than tomorrow but I cannot say how much more likely with any precision.
- Unknown probabilities - the domain of irreducible uncertainty. There is non-comparable premises. We cannot say what the outcome of the Iraq War. We cannot say what type of bank regulation will come out of Congress.
"It depends on the confidence with which we make this forecast or how likely we rate the likelihood of our best forecast turning out quite wrong. The state of confidence as they term it is a matter to which practical men always pay the closet attention."
Uncertainty will affect the impact of any statistical work.
"With a free hand to choose coefficient and time lags one can with enough industry always cook a formula to fit moderately well a limited range of past facts. But what does that prove?"
From Robert Skidelsky, Keynes: The Return of the Master