Immortality & Bayesian Statistics

Jabba

Jabba

Member
#61
#61
- So anyway, at this point, I’m thinking that my claim about immortality needs some re-wording.

1. Still figure that P(WEBR) + P(~WEBR) = 1.00.
2. And, that P(WEBR) = .99.
3. So, P(~WEBR) = .01

4. And, L(me|WEBR) is virtually zero…

5. Now, for the moment, assume that “~WEBR” can only be immortality.
6. Under these conditions, I claim that the entirety of P(~WEBR) must be accounted for by L(me|WEBR) + L(me|~WEBR).
7. And, we don’t need to do our usual estimating of L(me|~WEBR).
8. We can simply subtract L(me|WEBR) from 1.00.
9. And, under these conditions, since L(me|WEBR) is virtually zero, L(me|~WEBR) is virtually 100%.

10. Hopefully, #6 is our major sticking point so far.
11. I have more to say – but should probably see if anyone agrees with me so far.
 
bryangoodrich

bryangoodrich

Probably A Mammal
#62
#62
Jabba said:
6. Under these conditions, I claim that the entirety of P(~WEBR) must be accounted for by L(me|WEBR) + L(me|~WEBR).
Click to expand...
Claim it all you want, it doesn't make it correct. Mathematically, there is no reason for that statement to be true. If you want to simply insist it to be true for your conclusion to be validated, then you might as well forgo this charade and just say you claim your conclusion to be true. The math does not help you.
 
Jabba

Jabba

Member
#63
#63
- Here’s what I think is going on in our hypothetical situation.

- It is decided that there are only two possible ways to account for something (me).
- We think that one of those ways is much more probable than the other, and accord it a 99% (prior) probability.
- We naturally accord the second way a 1% (prior) probability.
- But then, based upon new evidence, we decide that the probability of the first way to account for me is really just a little better than zero percent.
- If we still think that there are only the two possible ways to account for me, we need to re-calculate the probability of the second way.
- “Normal” logic would now lead us to conclude that the probability of the second way accounting for me is just a little less than one hundred percent…

- However … apparently, the Bayesian model doesn’t allow for us to use the “normal” logic…

- I say -- OK, but by deciding that the first way of being responsible has such a small (prior) probability, we certainly do not decrease the (prior) probability of the second way.
- And, we should allow that the second way still has at least one percent (prior) probability of being responsible.
- If we use that figure for L(me|~WEBR), and virtually zero percent for L(me|WEBR), the posterior probability of WEBR becomes virtually zero, while the posterior probability of ~WEBR becomes virtually 1.00.



- P(me|WEBR) = L(me|WEBR)*P(WEBR)/(L(me|WEBR)*P(WEBR)+ L(me|~WEBR)*P(~WEBR)),
- Or, P(me|WEBR) = 1/10100!*.99/(1/10100!*.99)+ .01*.01.
- Or, virtually zero.
- And, P(me|~WEBR) = (1/10100!*.99)+ .01*.01./ (1/10100!*.99).
- Or, virtually 1.00.
 
Dason

Dason

Ambassador to the humans
#64
#64
Jabba said:
- “Normal” logic would now lead us to conclude that the probability of the second way accounting for me is just a little less than one hundred percent…

- However … apparently, the Bayesian model doesn’t allow for us to use the “normal”
Click to expand...
That's because "normal" logic has no basis in truth.

You have a bag with 999 dice that have 10000 sides and 1 other die with an unknown number of sides. You pull a die out and see the number 10000. Th probability of seeing 10000 from one of the first type of die is 1/10000. So according to your "normal" logic we most likely grabbed the other die. Do you see how silly this is? You know nothing about the probability of the other die showing 10000 - you literally can't deduce this probability using just what you know which is what we have been telling you. If you keep claiming that you can then you're delusional.

By the way the other die only had six sides and it was impossible to see 10000. Now wouldn't you look silly if you declared that you grabbed the "other" die.
 
Jabba

Jabba

Member
#65
#65
Dason said:
That's because "normal" logic has no basis in truth.

You have a bag with 999 dice that have 10000 sides and 1 other die with an unknown number of sides. You pull a die out and see the number 10000. Th probability of seeing 10000 from one of the first type of die is 1/10000. So according to your "normal" logic we most likely grabbed the other die. Do you see how silly this is? You know nothing about the probability of the other die showing 10000 - you literally can't deduce this probability using just what you know which is what we have been telling you. If you keep claiming that you can then you're delusional.

By the way the other die only had six sides and it was impossible to see 10000. Now wouldn't you look silly if you declared that you grabbed the "other" die.
Click to expand...
Dason,
- That's a different issue. You're talking about L(me|WEBR). I'm talking about L(me|~WEBR).
- I think that in order to clear up real disagreements, we need to deal with just one at a time. For now, I'll assume that you guys no longer have an issue with my choice for L(me|~WEBR). I'll be back shortly to discuss L(me|WEBR) -- unless someone wants to continue the L(me|~WEBR) discussion first.
 
Jabba

Jabba

Member
#66
#66
Dason,
- Here's my previous explanation for L(me|WEBR).

64. But then, we notice an “escape clause” (our other major doubt) for still using 100% as the likelihood of my current existence…
65. We realize, or remember, that specifically unlikely things have to happen.” E.g.,
a. Somebody has to win the lottery.
b. Somebody has to win the critical poker hand.
66. And consequently, we shouldn’t be basing our conclusions upon the likelihood of a specific outcome (me) in these cases – we should treat the specific case as someone/anyone.
67. And, the likelihood of someone/anyone eventually winning is essentially 100%.
68. And, we accept that the winners were just lucky.
69. And, we’re right – most of the time…

70. But what if the poker winner is a friend of the dealer, and they have been separately suspected of cheating in the past?
71. Comparing the lottery situation to the poker situation,
a. Knowing what I do about the lottery system, I’m pretty much stuck with luck as the explanation.
b. But, for the poker situation, there is another possible – and, possibly more probable -- explanatory model than luck…
72. I can reasonably eliminate cheating as an explanation in the lottery situation; but, I cannot do the same in the poker situation.
73. In the case of my own existence, I cannot reasonably eliminate the possibility that our explanatory model is wrong – and, in other words, that I do not have but one, finite, life to live.
74. That’s the key.
 
Dason

Dason

Ambassador to the humans
#67
#67
I'm done. You refuse to address the issues we bring up. I didn't want to call you a troll and as a mod I probably shouldn't resort to it. But you're acting like a troll by refusing to learn from and address the issues we raise. I didn't say anything about your specific probabilities in my last point. I was showing you why your "normal" logic is worthless and has no place here. You can't dismiss that by saying I was talking about L(me | WEBR) because I WASNT. You need to learn about probability. Specifically the very things we've pointed out many times already. If you're going to post here about probability you can't ignore us when we tell you you're doing it wrong. I've been patient up until now. Clean up your act and actually make an effort to understand and address our points. I'll keep monitoring this thread to make sure forum rules are followed but I'm done posting until you accept that you need to take the time to understand basic probability arguments.
 
Last edited:
victorxstc

victorxstc

Pirate
#71
#71
I agree with this one:

Jabba said:
73. In the case of my own existence, I cannot reasonably eliminate the possibility that our explanatory model is wrong – and, in other words, that I do not have but one, finite, life to live.
Click to expand...
You successfully proved that we will live after death and that soul and reincarnation exist :eek: Well done and thanks.
 
bryangoodrich

bryangoodrich

Probably A Mammal
#72
#72
Are you joking? Nobody has taken you seriously here when every response you have is a dogmatic belief that you're right to the point you'll butcher basic mathematics. Mathematics aside, your reasoning is shoddy and your concepts simplistic. All of this I've said already, so the fact you seem perplexed can only mean to me that you had no idea what I was talking about.
 
spunky

spunky

Can't make spagetti
#73
#73
bryangoodrich said:
Nobody has taken you seriously here when every response you have is a dogmatic belief that you're right to the point you'll butcher basic mathematics.
Click to expand...
what he said.

especially the "butchering basic mathematics" part. it's painfully obvious that you're using concepts you barely understand to try and make grandiose claims that are, by any standard, unscientific.

(notice that i did not say they could not be true. i am just saying they are unscientific and, therefore, a forum dedicated to philosophy or theology would probably be better suited for them)
 
CB

CB

Super Moderator
#74
#74
Jabba said:
CB and Bryan,
- Do you agree with Dason?
Click to expand...
Yes. You posted here presumably because you have some awareness that you don't know much about probability. But you seem to be unwilling to listen to anything that any of us are saying about probability. Instead you just keep reiterating the same, obviously flawed, arguments. This has become a waste of time.

FYI for others, there are some incredibly long discussions on this same topic, initiated by jabba, in the international skeptics forum:

http://www.internationalskeptics.com/forums/showthread.php?t=248163
http://www.internationalskeptics.com/forums/showthread.php?t=279595
http://www.internationalskeptics.com/forums/showthread.php?t=272394
 
Dason

Dason

Ambassador to the humans
#75
#75
Nice find CB. I don't want to lock this thread but just an FYI for anybody that feels like contributing more to this thread... One of those threads linked to has 193 pages - another has 75. It seems that jabba's main issue is that he has a conclusion he wants to "prove" and is grasping at straws trying to find any way possible to convince others that his proof is correct. Now if he has legitimate probability questions I would be more than happy to discuss them but seeing how much those other threads got dragged out I am not willing to continue on a discussion on this particular matter at all.
 
victorxstc

victorxstc

Pirate
#76
#76
Originally Posted by Giordano

You realize that Jabba only responds to what appears to be the least critical post, then still ignores the real concern in that post but tries to build a fake consensus around it?




I don't think there is any question about it.

Unless I'm mistaken, he strutted like a barnyard rooster in this very thread, claimed victory and offered to take the Million Dollar Challenge because his interlocutors had grown weary of arguing with a brick wall.
Click to expand...
http://www.internationalskeptics.com/forums/showpost.php?p=9880534&postcount=4001
 
victorxstc

victorxstc

Pirate
#79
#79
I'm becoming more and more positive that jabba is nothing but a computer program... It is sensitive only to certain patterns, and does not "see" the rest. If it ignores many posts, it is not because it is dodging them, but it can't literally identify and 'see' them since they are not pre-defined in its knowledge-based module.

Besides, its responses are limited to a list of specific answers and no "creativity" is really evident. It uses few random and irrelevant extra answers to make it look more humane, but they are too few and too irrelevant.

I congratulate its programmers for passing the Turing test for so long. However, they should work on its "learning module" more, as it appears only to repeating a certain pattern of answers (with few random additions), and does not really "learn".
 
You must log in or register to reply here.