Determining Probability

There’s been some debate here at TSZ recently about probability and the interpretation of probability.

I took some flak (my personal subjective opinion) for attempting to distinguish between calculating probabilities and estimating probabilities.

Yet in recent reading I came across this bit of text:

How do you determine the probability that a given event will occur? There are two ways: You can calculate it theoretically, or you can estimate it experimentally by performing a large number of trials.

– Probability: For the Enthusiastic Beginning. p. 335

Are there there ways to determine the probability that an event will occur that do not fall under wither of the two stated above, and if so, do any of those alternative ways to determine the probability that an event will occur involve neither calculating nor estimating the probability?

Bonus award if you can explain “epistemic probability” why it is 1) frequentist, or 2) not frequentist. My own reading puts it in the latter.

22 thoughts on “Determining Probability

  1. So probabilties based on observation should be taken with a grain of salt!

    – p. 336

    LoL. I kid you not. And this from a book where the approach is admittedly frequentist.

    Good thing evolutionists don’t depend on this sort of reasoning. Because, you know, they are all Bayesians. Not.

  2. Phood for phoodoo:

    Of course, based on induction over the millennia, our view of the universe probably is way off base. But let’s not get into that here.

    – p. 337

    But how could that possibly be?

  3. Probability doesn’t work in the real universe.
    This because it starts out denying a God(s) who would interfere in events.
    Without this supernatural option then probability concepts are already skewed.
    Since one would not know how much a God would interfere then probability never makes sense in a true analysis.
    It only works in special cases where God is not involved. This presumed.

    In origin issues I only see probability mattering when , AFTER presuming mutations could created a new population, then they estmate how probability could make a rhino out of a newly landed air breathing fish.

    Another cute thing is about probability of who will , thoughtfully, believe in creationism or evolutionism.
    Since this is a intellectual subject I see conclusions on it as as more, probable, to lean toward the smarter peoples getting the right answer. Therefore the smarter peoples would not just get this subject right but others AND SO those who are creationists and those who are evolutionists would cluster , in many subjects, as groups.
    Creos more likely to be conservative and evos more likely to be liberal.

  4. It [probability] only works in special cases where God is not involved.

    Therefore, when probability works, God is not involved.

    Take that you evolutionists!

  5. Mung: Phood for phoodoo:

    Of course, based on induction over the millennia, our view of the universe probably is way off base. But let’s not get into that here.

    – p. 337

    But how could that possibly be?

    Interestingly, one theory of how you calculate how far off base you are is by taking your observations, noting how far off base they are then multiplying that times 10.4 .

    Its just a slightly modified version of keiths epistemic probability equation.

  6. phoodoo: Its just a slightly modified version of keiths epistemic probability equation.

    Even so, I’ll be forever grateful to keiths, who rightly pointed out that you may know something that I don’t know, and that this may give you a certain advantage if you can entice me into making an unfavorable wager.

    Because, you know, I’ve never done that myself. Wouldn’t even think of it. Honest Mung be my name.

  7. Teach Yourself Statistics. No coverage of “epistemic probability.” Probably just a statistical oversight.

  8. Statistics: A Self-Teaching Guide. No coverage of “epistemic probability.” Probably just another statistical oversight.

  9. Probability and Statistics. No coverage of “epistemic probability.” Probably just another statistical oversight.

  10. Now if keiths is right, Statistics for the Utterly Confused ought to show it. Epistemic probability. Not mentioned. No coverage of “epistemic probability.” Probably just another statistical oversight.

  11. Just like Rumraket didn’t understand that he was really talking about “epistemic probabilities,” all these books on statistics don’t understand that they are also really talking about “epistemic probabilities.”

    Under the covers. Secret language. Special handshakes.

  12. Probability for Dummies. No coverage of “epistemic probability.” Probably just another statistical oversight.

  13. Statistics for Dummies. No coverage of “epistemic probability.” Probably just another statistical oversight.

  14. I’m beginning to lose hope.

    Statistics Demystified. No coverage of “epistemic probability.” Probably just another statistical oversight.

  15. Statistics: The Easy Way. No coverage of “epistemic probability.” Probably just another statistical oversight.

  16. Joe Felsenstein: Try looking also under “subjective probability”. This Wikipedia page may help too.

    Oh, I can find “epsitemic probability” online no problem, lol. Thanks for the tip though.

    I’ll try looking for subjective probability. I think I did come across that in one book, but iirc it’s Bayesian, not frequentist. I’ll need to double check that.

  17. From a different thread:

    keiths: Rhetorical question: Why do so few IDers understand the basics of evolution and probability/statistics, when those are precisely the subjects upon which their (flimsy) arguments depend?

    ok, sure. it was a rhetorical question. But it does bring up a point that I can agree with keiths about. To the extent that ID arguments rely on probabilistic reasoning, there’s been a poor job of teaching the material necessary to understand those arguments. Definite room for improvement there.

    It also brings up something that I disagree with keiths about. Since most people simply don’t understand the math of DEM, most people don’t rely on it. So when keiths claims ID arguments depend on probability/statistics, which arguments does he have in mind?

Leave a Reply