Recent Wars and Bottlenecks

[LeBrok]

Can we speak of the Jewish Holocaust as an example of bottleneck in Europe?

Bottle necking will happen when there is just handful of people left. Statistics of small numbers.

 

[castelleone]

Check statistics of big numbers.

Thank you.
But that neither excludes the possibility of an haplogroup become extinct (as a result of the war) nor allows us to speak of a reduction or increase of haplogroup distribution at the exact same rate.

 

[castelleone]

Bottle necking will happen when there is just handful of people left. Statistics of small numbers.

Thank you.
I understand statistics of small numbers as some sort of fallacious argument. Is that what you mean?

 

[LeBrok]

Thank you.
I understand statistics of small numbers as some sort of fallacious argument. Is that what you mean?

Fallacious argument?!!! Are you sure you understand statistics and probabilities? Because this is the key here.

Take a dice. Imagine that every one of these numbers represent a haplogroup. Now, let's deal with small tribe of 12 males, 6 haplogroups, 2 each. They went to the battle and half of them died. Throw a dice 6 times to see what haplogroups will be eliminated. Probably one or two haplogroup was totally eliminated, the rest of proportions are off too.
Now, repeat same experiments for 120 or much better for 1,200 people tribe, 6 haplogroups, 200 people in each haplogroup. You will see that it will become impossible to kill off one haplogroup completely, or even change haplogroups proportions.
Learn this and you will know how bottleneck works.

 

[castelleone]

Fallacious argument?!!! Are you sure you understand statistics and probabilities? Because this is the key here.

Take a dice. Imagine that every one of these numbers represent a haplogroup. Now, let's deal with small tribe of 12 males, 6 haplogroups, 2 each. They went to the battle and half of them died. Throw a dice 6 times to see what haplogroups will be eliminated. Probably one or two haplogroup was totally eliminated, the rest of proportions are off too.
Now, repeat same experiments for 120 or much better for 1,200 people tribe, 6 haplogroups, 200 people in each haplogroup. You will see that it will become impossible to kill off one haplogroup completely, or even change haplogroups proportions.
Learn this and you will know how bottleneck works.

Thank you.
But please, dont overreact, it is neither polite nor helpful and makes you loose credibility.


There is a book called 'The Daemon Haunted World', which describes many fallacies, one of which is called 'statistics of small numbers'.
That is all I meant.

May I suggest you read 'The Black Swan' by Nassim Taleb?
He dedicates a whole chapter on the foolish notion of seeing all sorts of circumstances as dice experiments.
He uses economy, you might apply it to wars.

You can learn something from him:
https://www.youtube.com/watch?v=nDY_fh2TVlI

And it is too difficult for you, then you can learn from this guy:
https://youtu.be/SgRzy862zkc?t=2m8s

 

[LeBrok]

Thank you.
But please, dont overreact, it is neither polite nor helpful and makes you loose credibility, ok?


There is a book called 'The Daemon Haunted World', which describes many fallacies, one of which is called 'statistics of small numbers'.
That is all I meant.

May I suggest you read 'The Black Swan' by Nassim Taleb?
He dedicates a whole chapter on the foolish notion of seeing all sorts of circumstances as dice experiments.
He uses economy, you might apply it to wars.

You can learn something from him:
https://www.youtube.com/watch?v=nDY_fh2TVlI

And it is too difficult for you, then you can learn from this guy:
https://youtu.be/SgRzy862zkc?t=2m8s

Whatever dude, did you get the statistical probabilities I explained? Do you understand now why bottleneck can only happen in very small groups?
To understand this you don't need to read a book even. Just do the experiment I described. Do the science yourself!

 

[castelleone]

Whatever dude, did you get the statistical probabilities I explained? Do you understand now why bottleneck can only happen in very small groups?
To understand this you don't need to read a book even. Just do the experiment I described. Do the science yourself!

Thank you.
Please don't use expressions like 'Whatever dude...', because it makes you sound immature and uneducated.

Again, the experiment which you described is indeed the textbook example why dice (and coins) does not work in some statistical domains ruled by different probability distribution functions. I am therefore totally convincend that you didn't understand where the law of large numbers can be applied and where it cannot, therefore you dismiss it.

Mr. Taleb has also written a book called 'Fooled by Randomness', which deals with the fact that many people view the world as absolutely measurable and predictable (gaussian-modelled) by appling statistics in the most foolish ways. I suggest you educate yourself by reading not only one, but both books.

 

[mwauthy]

Can we speak of the Jewish Holocaust as an example of bottleneck in Europe?

I think another less recent example would the the distribution of y haplogroups in North America before and after the arrival of Europeans, and the impact disease had on Native Americans.

The Genghis Khan genocide on the Tangut people is another example of warfare affecting haplogroup frequencies.

 

[LeBrok]

Thank you.
Please don't use expressions like 'Whatever dude...', because it makes you sound immature and uneducated.

Again, the experiment which you described is indeed the textbook example why dice (and coins) does not work in some statistical domains ruled by different probability distribution functions. I am therefore totally convincend that you didn't understand where the law of large numbers can be applied and where it cannot, therefore you dismiss it.

Mr. Taleb has also written a book called 'Fooled by Randomness', which deals with the fact that many people view the world as absolutely measurable and predictable (gaussian-modelled) by appling statistics in the most foolish ways. I suggest you educate yourself by reading not only one, but both books.

One last time, let's try a simple logic. Small tribes = small numbers statistics/law. Big tribes = big number statistics/law. Don't apply large numbers law to small tribes!!! Small tribe and small number law explain bottleneck phenomenon very easily, therefore it is the right way, it is the true measure and explanation.
Again, you don't need to get confused by some books. Do the experiment I described and bottleneck will become understandable. Don't be stubborn and contrarian and do this simple experiment.

 

[Angela]

The impact of disease may be a separate category if the carriers of certain haplogroups lack the needed immunity genes in their autosomal dna.

 

[castelleone]

One last time, let's try a simple logic. Small tribes = small numbers statistics/law. Big tribes = big number statistics/law. Don't apply large numbers law to small tribes!!! Small tribe and small number law explain bottleneck phenomenon very easily, therefore it is the right way, it is the true measure and explanation.
Again, you don't need to get confused by some books. Do the experiment I described and bottleneck will become understandable. Don't be stubborn and contrarian and do this simple experiment.

Thank you.
Please don't loose your patience, it makes you sound intelectually insecure.

[ 1 ] Killing will affect all haplogroups proportionally equally in huge populations.
[ 2 ] Nope, for big numbers of population all the haplogroups will be reduced by x%. Ratio of haplogroups to each other will not change.

Wrong. If and only if the stochastic process is modelled by a specific probability distribution function (and you must be very lucky to find even a semi-perfect match!) and is repeated infinitely so that time averaging (evening out of rates...) takes places. This is not the case! Abrupt events like wars, that come as a surprise, have a major effect, and are often inappropriately rationalized after the fact with the benefit of hindsight, do happen.

[ 3 ]Do you understand now why bottleneck can only happen in very small groups?

Wrong. You assume probability of bottleneck for large populations is p = 0 (it doesn't happen), when it is actually p > 0 (it can happen). In other words, it is not impossible, but probable, that bottlenecks can and will happen in any population of any size as a result of war. Even if you lack higher knowledge in statistics, it is never to late to learn how you are fooled by randomness of the particular events involved.

[ 4 ]Now, repeat same experiments for 120 or much better for 1,200 people tribe, 6 haplogroups, 200 people in each haplogroup. You will see that it will become impossible to kill off one haplogroup completely, or even change haplogroups proportions.

Wrong. You insist on using an example of dice to support your argument, despite that literature already dismissed this approach a long time ago. "One problem, labeled the ludic fallacy by Taleb, is the belief that the unstructured randomness found in life resembles the structured randomness found in games." Source: https://en.wikipedia.org/wiki/Black_swan_theory.

 

[castelleone]

The impact of disease may be a separate category if the carriers of certain haplogroups lack the needed immunity genes in their autosomal dna.

There is a book called 'Guns, Germs and Steel' that brings substantial information on the topic.
https://en.wikipedia.org/wiki/Guns,_Germs,_and_Steel

 

[LeBrok]

Thank you.
Please don't loose your patience, it makes you sound intelectually insecure.




Wrong. If and only if the stochastic process is modelled by a specific probability distribution function (and you must be very lucky to find even a semi-perfect match!) and is repeated infinitely so that time averaging (evening out of rates...) takes places. This is not the case! Abrupt events like wars, that come as a surprise, have a major effect, and are often inappropriately rationalized after the fact with the benefit of hindsight, do happen.



Wrong. You assume probability of bottleneck for large populations is p = 0 (it doesn't happen), when it is actually p > 0 (it can happen). In other words, it is not impossible, but probable, that bottlenecks can and will happen in any population of any size as a result of war. Even if you lack higher knowledge in statistics, it is never to late to learn how you are fooled by randomness of the particular events involved.



Wrong. You insist on using an example of dice to support your argument, despite that literature already dismissed this approach a long time ago. "One problem, labeled the ludic fallacy by Taleb, is the belief that the unstructured randomness found in life resembles the structured randomness found in games." Source: https://en.wikipedia.org/wiki/Black_swan_theory.

Well, if you don't want to "sacrifice" an hour of your time to do a simple experiment, do science yourself, which would give you a total explanation of bottleneck effect, then stay confused about this subject forever.

 

[LeBrok]

Wrong. You assume probability of bottleneck for large populations is p = 0 (it doesn't happen), when it is actually p > 0 (it can happen).

What is the probability of bottleneck effect (let's say losing one haplogroup out of 10) happening to population shrinking from 10 million to 1 million, due to a war let's say? Astronomical! You are welcome to do the math.

What is probability of losing one haplogroup in population shrinking from 100 to 10 people. It will happen every time, almost.

Of course if 1 million population shrinks down to 100 or 10 people the bottleneck effect will still happen. Though probability of large group shrinking to just few is much smaller than small group shrinking to few. At the end, in both cases, we have to deal with very small remaining group to see bottleneck effect. And the smaller the remaining group the bigger the effect.

 

[castelleone]

What is the probability of bottleneck effect (let's say losing one haplogroup out of 10) happening to population shrinking from 10 million to 1 million, due to a war let's say? Astronomical! You are welcome to do the math.

It is hilarious...
Before you said the bottleneck won't happen, now you admit it can happen and even foolishly state that the probability of it happening is enormously or inconceivably large or great. You are welcome to consult a dictionary.

What is probability of losing one haplogroup in population shrinking from 100 to 10 people. It will happen every time, almost.

When did I say it won't happen?

Of course if 1 million population shrinks down to 100 or 10 people the bottleneck effect will still happen. Though probability of large group shrinking to just few is much smaller than small group shrinking to few. At the end, in both cases, we have to deal with very small remaining group to see bottleneck effect. And the smaller the remaining group the bigger the effect.

... and now you admit it will happen from 1 million to 10.

By the way, next time you suggest an experiment, don't forget to drop some imaginary nuclear bomb in London.

I rest my case.

 

[castelleone]

Well, if you don't want to "sacrifice" an hour of your time to do a simple experiment, do science yourself, which would give you a total explanation of bottleneck effect, then stay confused about this subject forever.

Thank you.
I won't sacrifice an hour of my time doing an experiment which is nothing more than (pardon me the language) intellectual masturbation.

 

[castelleone]

I think another less recent example would the the distribution of y haplogroups in North America before and after the arrival of Europeans, and the impact disease had on Native Americans. ...

"Although historical evidence regarding population size during the 16th century is fragmentary, surviving personal and governmental accounts document substantial, if not catastrophic, declines during the period. ... The signature of the bottleneck is also apparent in the genealogical structure of the dataset. A majority-rule consensus genealogy constructed from the trees sampled by the Markov chain demonstrates a transiently increased rate of coalescence near the tips of the genealogy (Fig. 4), particularly in haplogroups C and D. Because the rate at which lineages coalesce is inversely proportional to population size, the preponderance of coalescent events near the tips in these haplogroups suggests that the bottleneck affected haplogroups C and D disproportionately."

Here it is important to pause, smile and reflect about some deterministic statements made in this thread, more specifically the following one "for big numbers of population all the haplogroups will be reduced by x%. Ratio of haplogroups to each other will not change."

"... our results are consistent with historical records suggesting that epidemics, warfare, enslavement, and famines resulted in significant population declines among Native Americans during the 16th century. Additionally, the scale of the contraction suggests that the depopulation was not localized to particular regions or communities, and instead, was likely to have been widespread or to have had an especially severe impact on the most populous regions." Source:Native Americans experienced a strong population bottleneck coincident with European contact

Here it is important to pause, smile once more and reflect about yet another deterministic statement made in this thread, more specifically the following one "bottleneck can only happen in very small group."