Interpretering Bayesian Analysis

The catch-all forum for general topics and debates. Minimal moderation. Rated PG to PG-13.
User avatar
Dr Moore
Endowed Chair of Historical Innovation
Posts: 1812
Joined: Mon Oct 26, 2020 2:16 pm
Location: Cassius University

Re: Interpretering Bayesian Analysis

Post by Dr Moore »

Analytics wrote:
Mon Jul 26, 2021 8:46 pm
I would say that in theory, Bayesian analysis encourages you to ask the right questions, makes your arguments more explicit, and if you know what you are doing, weight them correctly.
QFT emphasis mine.
Analytics wrote: Whether you are a Texas sharpshooter depends upon your assumptions. When Kyler says there is a 0.0001% chance that a fictional book would be written in 65 days but a 100% chance that a historical book would be, the second half of that is the Texas sharpshooting. If somebody else were to reply that the chances are 0.00001% for a fictional book to be written in 65 days and the chances are 0.00001% for a historical book to be written in 65 days, he is not Texas sharpshooting.
So much more than this, but that's a good start.

(1) What are the chances that a practiced fictional story-teller (treasure digs) turned Biblical story teller (Methodist exhorter) will spend XX number of years concocting, plus YY days to dictate Bible fan fiction, for which there is virtually no comparable product, and subsequently not bother with careful editing? Is it really 0.00001% or is the better answer that "For this prior, we have no idea because we have insufficient data to construct a control group and the book itself is a singular work of art by a singular artist.

(2) What are the chances that a historical book could be transcribed in 65 days by miraculous means? Well, ok then. It should be 100% - right? Let's go with that, but this assumption also exposes the evidence point to a separate qualifying Bayesian test. If words are read from a stone, written, read back and then verified, what is the "penalty" to the miraculous claim if 1 mistake is made? What is the "penalty" to the miraculous claim if 1.5 mistakes per 150 words are made? And if anachronisms appear, what is the "penalty" for 1 anachronism? What is the penalty for 16th or 19th century anachronistic phraseology appearing? And yes, every anachronism must be evaluated separately. And moreover, what is the penalty for having *both* transcription errors AND anachronisms?

So for (2) above, one option is amateur sharpshooter hour, assuming chances at 100% for dictation of a truly ancient text. But to be serious, the Mopologists have to apply rigor on that assumption, which means they have to do something they don't want to do -- as Doctor Scratch suggested earlier. Which is to properly consider evidence of tight and loose translation theories on their merits and as mutually exclusive criteria.

As Analytics points out here, the net outcome is that the gap between authenticity and forgery will close. Instead of 1% vs 0.00001%, we may end up with two equally remote scenarios, which means Kyler's study, if done seriously, would have to conclude that the dictation process is indistinguishable from fraud.
User avatar
Gadianton
God
Posts: 3843
Joined: Sun Oct 25, 2020 11:56 pm
Location: Elsewhere

Re: Interpretering Bayesian Analysis

Post by Gadianton »

Analytics wrote:When Kyler says there is a 0.0001% chance that a fictional book would be written in 65 days but a 100% chance that a historical book would be, the second half of that is the Texas sharpshooting. If somebody else were to reply that the chances are 0.00001% for a fictional book to be written in 65 days and the chances are 0.00001% for a historical book to be written in 65 days, he is not Texas sharpshooting.
hmm. Is sharpshooting relative, or do you call it something else if the main power thrusters are in the microscopic odds of the alt-hypothesis?

Using the numbers Kyle provided, except substituting a 50-50 prior, even if we say that there is a .000001 chance that a historical book would be written in 65 days, the chances of a fictional book being written in 65 days are so much lower that the probability the Book of Mormon is historical is still 99.7%
User avatar
Doctor CamNC4Me
God
Posts: 8981
Joined: Wed Oct 28, 2020 2:04 am

Re: Interpretering Bayesian Analysis

Post by Doctor CamNC4Me »

On the Road was written by Jack Kerouac in three weeks. I don't see why Joseph Smith, if he had the framework of the narrative worked out, was willing to 'borrow' heavily from other books and the Bible, and had a school teacher as the primary scribe (basically demonstrating literacy, education, and competency), couldn't produce something along the lines of the Book of Mormon manuscript, which was in dire need of editing, anyway. It wouldn't really be a thing to have source material on hand, that they could review and cherry pick when needed. Remember, the short 'translation' period is a story of their invention, like everything else associated with early Mormonism, so I wouldn't get too hung up on how long it supposedly took.

But. If the timeline is a Bayesian thing, then Dr. Rasmussen really should only take into account original material found in the Book of Mormon and not anything that could be lifted from other books. AND, On the Road might be a good metric as a possibility to produce a book in short order.

- Doc
Hugh Nibley claimed he bumped into Adolf Hitler, Albert Einstein, Winston Churchill, Gertrude Stein, and the Grand Duke Vladimir Romanoff. Dishonesty is baked into Mormonism.
User avatar
Doctor CamNC4Me
God
Posts: 8981
Joined: Wed Oct 28, 2020 2:04 am

Re: Interpretering Bayesian Analysis

Post by Doctor CamNC4Me »

Now that I'm doing some quick Google searches, I read that the Strange Case of Dr Jekyll and Mr Hyde by Robert Louis Stevenson was written in three to six days. Good Lord.

I suppose the 'out' someone might have with regard to the Book of Mormon is that the Mormon church doesn't consider it fiction, but c'mon, until it can actually be correlated with real, actual history it's fiction.

- Doc
Hugh Nibley claimed he bumped into Adolf Hitler, Albert Einstein, Winston Churchill, Gertrude Stein, and the Grand Duke Vladimir Romanoff. Dishonesty is baked into Mormonism.
Dr Exiled
God
Posts: 1602
Joined: Tue Oct 27, 2020 2:40 pm

Re: Interpretering Bayesian Analysis

Post by Dr Exiled »

Where does he come up with these numbers anyway? Looks like they are taken right out of his _____.

Why not 0.001% and 0.0001% or how about 0.0002% and 0.00002% respectively? How about putting an 8 after the zeros instead of a 1 or 2? It seems that all we can say is that the chances are small that a fictional or historical book was written in the disputed time period, but to assign an actual number seems to be too much guesswork to even be a valid exercise. Is there a function to account for the fuzziness of history?
Myth is misused by the powerful to subjugate the masses all too often.
drumdude
God
Posts: 5219
Joined: Thu Oct 29, 2020 5:29 am

Re: Interpretering Bayesian Analysis

Post by drumdude »

Dr Exiled wrote:
Tue Jul 27, 2021 12:51 am
Where does he come up with these numbers anyway? Looks like they are taken right out of his _____.

Why not 0.001% and 0.0001% or how about 0.0002% and 0.00002% respectively? How about putting an 8 after the zeros instead of a 1 or 2? It seems that all we can say is that the chances are small that a fictional or historical book was written in the disputed time period, but to assign an actual number seems to be too much guesswork to even be a valid exercise. Is there a function to account for the fuzziness of history?
What makes me laugh is how many significant figures he puts into them. We're talking about guesses here, but he's using 6 significant figures as if he's measuring something in a lab!

Probability of Joseph being tired and falling into his hat: 0.25582203% of course!
huckelberry
God
Posts: 2579
Joined: Wed Oct 28, 2020 3:48 pm

Re: Interpretering Bayesian Analysis

Post by huckelberry »

chance of Joseph writing out and memorizing the Book of Mormon, .00000000002 the chance he developed the general story cycle in rough form verbally of a period of years and then spoke like a story teller, .74

or somethings close to that. I do not see an exact method for measurement.
Analytics
Elder
Posts: 350
Joined: Wed Oct 28, 2020 3:11 pm

Re: Interpretering Bayesian Analysis

Post by Analytics »

Gadianton wrote:
Mon Jul 26, 2021 10:09 pm
Analytics wrote:When Kyler says there is a 0.0001% chance that a fictional book would be written in 65 days but a 100% chance that a historical book would be, the second half of that is the Texas sharpshooting. If somebody else were to reply that the chances are 0.00001% for a fictional book to be written in 65 days and the chances are 0.00001% for a historical book to be written in 65 days, he is not Texas sharpshooting.
hmm. Is sharpshooting relative, or do you call it something else if the main power thrusters are in the microscopic odds of the alt-hypothesis?

Using the numbers Kyle provided, except substituting a 50-50 prior, even if we say that there is a .000001 chance that a historical book would be written in 65 days, the chances of a fictional book being written in 65 days are so much lower that the probability the Book of Mormon is historical is still 99.7%
Here is how I see it.

If we started with a 50-50 prior, what is the posterior probability if the chances of a fake book being written in only 65 days is 0.000001%?

It depends.

If the chances of an authentic book being written in only 65 days is also 0.000001%, then the posterior probability is still 50-50.

When Kyler said the chances of an authentic book being written in 65 days is 100%, he was drawing a tight bullseye around 65 days. Did he do that before he knew it was translated in 65 days, or after?
Analytics
Elder
Posts: 350
Joined: Wed Oct 28, 2020 3:11 pm

Re: Interpretering Bayesian Analysis

Post by Analytics »

drumdude wrote:
Tue Jul 27, 2021 2:42 am
Dr Exiled wrote:
Tue Jul 27, 2021 12:51 am
Where does he come up with these numbers anyway? Looks like they are taken right out of his _____.

Why not 0.001% and 0.0001% or how about 0.0002% and 0.00002% respectively? How about putting an 8 after the zeros instead of a 1 or 2? It seems that all we can say is that the chances are small that a fictional or historical book was written in the disputed time period, but to assign an actual number seems to be too much guesswork to even be a valid exercise. Is there a function to account for the fuzziness of history?
What makes me laugh is how many significant figures he puts into them. We're talking about guesses here, but he's using 6 significant figures as if he's measuring something in a lab!

Probability of Joseph being tired and falling into his hat: 0.25582203% of course!
Is Kyler really an actuary?

An actuary is a person who passes as an expert on the basis of a prolific ability to produce an infinite variety of incomprehensive figures calculated with micrometric precision from the vaguest of assumptions based on debatable evidence from inconclusive data derived by persons of questionable reliability for the sole purpose of confusing an already hopelessly befuddled group of persons who never read the statistics anyway.
dastardly stem
God
Posts: 2259
Joined: Tue Nov 03, 2020 2:38 pm

Re: Interpretering Bayesian Analysis

Post by dastardly stem »

Dr Rasmussen's previous episodes didn't make any sense to me. The length of the book somehow addresses the "is the Book of Mormon ancient or written in the 19th C"? I don't think there's any real connection. The consequent probability made no sense for episode 1. Episode 2 was worse in that whether or not Joseph dreamed or visioned God and Jesus has no applicability at all. But it came out as a negative for him, so I guess he's happy. Episode 3 was much like 1. The book exists. If he dictated it in 65 days, great. I mean amazing. But that has no bearing on whether it's ancient. Its more likely he perpetrated a trick on his scribes. It's more likely he collaborated with his scribes. It's more likely he wrote it over a number of years previous to getting it written down. ANd it's more likely he got the story from others. The episode didn't work for anything, if you ask me.

And now we have Episode 4

It actually makes sense. I hardly care to quibble with the numbers. But he created a valid looking binary evaluation. On the Hypothesis the book is ancient, what's the probability given our DNA understanding? VS on the Hyposthesis the book was written by Joseph Smith in the 19th Century, what's the probability given our DNA understanding.

I won't even quibble with the numbers even though I think they are kind of silly. I'm just happy we have an actual evaluation that makes sense.

It works out on the negative though.
“Every one of us is, in the cosmic perspective, precious. If a human disagrees with you, let him live. In a hundred billion galaxies, you will not find another.”
― Carl Sagan, Cosmos
Post Reply