|
Author
|
Topic: Universal probability bound?
|
RBH
Member
Member # 380
|
posted 12. December 2005 01:41
Irving wrote quote:
I suppose some things are known...
We know that sides is greater than 1. And we can assume that at least one side not equal to 4 (or the whole exercise is moot), and that the time required from beginning to end of the throw is not equal to zero. Thus regardless of the other factors, there is a finite limit to the number of throws possible in the allowable time... (bolding added)
The bolded assumption is not supported. We do not know how many of the sides have "4" on them. It may be zero or some number greater than zero. In any case, even assuming that at least one side does not have a "4" on it gets one no forarder in calculating the probability of a "4" coming up, since we still don't know how many sides there are, nor how many of the sides show "4", nor whether and how the die is loaded.
RBH
[ 12. December 2005, 01:41: Message edited by: RBH ]
IP: Logged
|
|
Irving
Member
Member # 535
|
posted 12. December 2005 07:16
But you may be missing the point.
I'm only illustrating that much can be known or inferred even without access to the die itself.
If I carefully observe the results of each roll, over time, I can better predict the outcome of the next roll. After thousands of rolls, I may become very accurate in predicting the probabilities of the next roll. In all cases I need not actually observe the die itself, though upon analysis of results I can begin to infer it's structure and nature.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 12. December 2005 16:47
Irving wrote quote:
But you may be missing the point.
I'm only illustrating that much can be known or inferred even without access to the die itself.
If I carefully observe the results of each roll, over time, I can better predict the outcome of the next roll. After thousands of rolls, I may become very accurate in predicting the probabilities of the next roll. In all cases I need not actually observe the die itself, though upon analysis of results I can begin to infer it's structure and nature.
Oh, sure, that's unproblematic -- that's what a whole slew of statistics is about: estimation. I have no quarrel with that. But my point is that absent those data, and absent any knowledge of the structure of the die, no probability estimation is possible. And it's the latter condition that characterizes all the probability doodah I see in creationist and ID writing.
RBH
IP: Logged
|
|
Irving
Member
Member # 535
|
posted 12. December 2005 19:54
Can't comment on Creationist probability writings...
Do you have a link?
The issue with ID however, is that we have observations and theories regarding how natural processes operate. That they operate in accordance with local environmental requirements. Therefore; while much is unknown, much can still be inferred by the observation of natural result. Thus there is some room to estimate probabilities of Natural processes mimicing the incorporation of "external" or non-locally derived requirements.
IP: Logged
|
|
Irving
Member
Member # 535
|
posted 12. December 2005 19:55
P.S. Again you may have missed the point. I do not need to KNOW the structure of the die. I can infer the structure by observation of the results...
IP: Logged
|
|
Atom
Member
Member # 1840
|
posted 18. December 2005 12:55
Hello all involved,
I have been reading this thread, and find it quite interesting and informative. I would also like to admit that the intelligences and knowledge displayed in this thread probably exceed mine, but I will try to contribute in what small fashion I can.
RBH brings up a valid point in calculating probabilities of things which we know little of their distributions. Melvin and Kyle, however, also raise interesting counter-points. In my reading of ID material, I have found that most of the calculations deal with discrete systems with geometric distributions, like dice. For example, I have seen calculations based on amino acid sequencing (discrete, 20 possibilities per location on the chain, and each AA can link to any other. Thus, the probability of finding AA1 at spot N+1 is independent of the the probability for space N.) and also on base sequences (Once again, for DNA/RNA bases you have a discrete set of possibile states for each spot on the chain, and the probability of base B being found on spot N+1 is independent of the previous base.) For this reason, I have found the probability calculations performed on these systems unproblematic. They basically perform as combinatoric/permutation spaces.
Maybe I'm missing some relevant finding showing that the probability of finding amino acid S on a polypeptide chain at spot N+1 is directly conditional on what amino acid precedes it; in which case, I'd be forced to acknowledge RBH's criticisms as very real indeed.
But to the best of my knowledge, the "letters" of DNA chains and polypeptide chains are not like english letters, which have rules about which can follow a previous letter. They seem to each be independent, like dice rolls. If I am wrong in this assertion, please forgive my ignorance.
IP: Logged
|
|
|
Printer-friendly view of this topic