Arp objects, QSOs, Statistics

[JEROME DA GNOME]

David, there is no good way to address the sampling issue. There is a huge dataset full of objects. Almost all of them obey Hubble's law, but there are a few anomalies. If you take one of those anomalies and ask, what's the probability this happened by chance, it will be very very small (that is what's called a posteriori statistics, and it's wrong and misleading). But if you only ask, what's the probability there will be some anomalies, it's basically 1.

Somewhere in between those two questions is the correct one to ask. The second question isn't satisfactory because it would lead you to ignore real interesting anomalies, but neither is the first, because it lends false significance to chance events.

In this case, the probability that the big bang model is wrong is ridiculously small. Nearly every object in the universe obeys a Hubble law, and anomalies are both expected and predicted from big bang theory. No object has precisely its Hubble velocity, and the differences are called peculiar velocities. A few of the billions of objects we see will have large peculiar velocities. So that's one possible explanation. Another is that they are wrong about how far away these things are. In astro measuring distance is extremely difficult, but without it you can't determine whether there's an anomaly (because Hubble relates distance to velocity, and hence redshift).

Furthermore the theory does a superb job explaining other observations too, such as the cosmic microwave background, it's consistent with particle physics, and it's predicted by general relativity (which we know independently is correct). There is no alternative theory that can explain those things.

How do you know that the redshift is correctly measuring time and distance?

 

[DeiRenDopa]

David, there is no good way to address the sampling issue. There is a huge dataset full of objects. Almost all of them obey Hubble's law, but there are a few anomalies. If you take one of those anomalies and ask, what's the probability this happened by chance, it will be very very small (that is what's called a posteriori statistics, and it's wrong and misleading). But if you only ask, what's the probability there will be some anomalies, it's basically 1.

Somewhere in between those two questions is the correct one to ask. The second question isn't satisfactory because it would lead you to ignore real interesting anomalies, but neither is the first, because it lends false significance to chance events.

In this case, the probability that the big bang model is wrong is ridiculously small. Nearly every object in the universe obeys a Hubble law, and anomalies are both expected and predicted from big bang theory. No object has precisely its Hubble velocity, and the differences are called peculiar velocities. A few of the billions of objects we see will have large peculiar velocities. So that's one possible explanation. Another is that they are wrong about how far away these things are. In astro measuring distance is extremely difficult, but without it you can't determine whether there's an anomaly (because Hubble relates distance to velocity, and hence redshift).

Furthermore the theory does a superb job explaining other observations too, such as the cosmic microwave background, it's consistent with particle physics, and it's predicted by general relativity (which we know independently is correct). There is no alternative theory that can explain those things.

How do you know that the redshift is correctly measuring time and distance?

Welcome to this thread, JEROME!

I'm puzzled though, what does your question have to do with the focus of this thread?

And on the question itself, "redshift" does not "measure time and distance"!

Would you like a brief introduction to the relevant physics and astronomy? If so, I'll start a new thread on just that topic.

 

[dudalb]

Welcome to this thread, JEROME!

Boy,are you gonna regret that.......

 

[DeiRenDopa]

Boy,are you gonna regret that.......

I hope not!

As BeAChooser has apparently left us, we have no one who can tell us how to do the calculations concerning the 'probability' of finding one configuration or another, and no one who can say if the results I got by applying what I think is BAC's approach (to calculating probabilities) is correct or not. I trust that JEROME can step up to the plate and provide some statistical meat to the poor bones we have to work with at the moment ...

 

[JEROME DA GNOME]

Welcome to this thread, JEROME!

I'm puzzled though, what does your question have to do with the focus of this thread?

Did you read what I quoted? It was refrencing Hubble's law.

And on the question itself, "redshift" does not "measure time and distance"!

You may want to read up on Hubble's law.

Would you like a brief introduction to the relevant physics and astronomy? If so, I'll start a new thread on just that topic.

You have it backwards.

 

[DeiRenDopa]

... snip ...

I'm puzzled though, what does your question have to do with the focus of this thread?

Did you read what I quoted? It was refrencing Hubble's law.

Indeed.

You did not answer my question - what has this got to do with the focus of this thread?

Please read the OP to get a good idea of what it's about - the focus is quite narrow (Arp, use of statistics, QSO/quasar-galaxy associations, that sort of thing).

And on the question itself, "redshift" does not "measure time and distance"!

You may want to read up on Hubble's law.

Would you like a brief introduction to the relevant physics and astronomy? If so, I'll start a new thread on just that topic.

You have it backwards.

Ah yes, I see now ...

OK, I'll start a new thread on the topic later today - I hope you can join!

ETA: done - it's called What is the observational evidence for the Hubble relationship?

 

[DeiRenDopa]

Set FOUR: 0.217353, 0.252693, 0.342904, 0.362706, 0.537829, 1.05378, 1.1014, 1.16468, 1.45072, 1.54642, 1.56929.

Here are the 'BAC probabilities' I calculate:

"Amaik peaks": 5.7x10^-7
"Karlsson peaks": 3.6x10^-5
"regular peaks": 5.7x10^-7
"DRDS peaks": 8.2x10^-8

Set FIVE: 0.365631, 0.7015, 0.746949, 0.937404, 0.963945, 1.03822, 1.1356, 1.2532, 1.37193, 1.73417, 1.8172, 1.86347, 2.1246, 2.3712.

Here are the 'BAC probabilities' I calculate:

"Amaik peaks": 1.9x10^-5
"Karlsson peaks": 2.3x10^-6
"regular peaks": 2.9x10^-9
"DRDS peaks": 8.0x10^-6
Now that BeAChooser has returned to posting in the JREF forum, I look forward to his confirming (or not) the correctness of these calculations ...

 

[Dancing David]

Hi Jerome!

Well, it looks as though the luminosity of Cephid variables is related to the redshift.

But Arp says he has objects where it would not be, based up a wishful assciation without any statistical meaning.

Bye jeroem, please return when you want to address how Arp's association has any more meaning than my finger pointing at the moon. they could run the representative samples any time that they want now.

Then DRD raises a great question in relation to this topic of: What is a QSO?

 

[Dancing David]

Now that BeAChooser has returned to posting in the JREF forum, I look forward to his confirming (or not) the correctness of these calculations ...

Not sure that will ever happen based upon past behavior.

 

[Tubbythin]

Hi Jerome!

Well, it looks as though the luminosity of Cephid variables is related to the redshift.

The variables have a period-luminosity relation (though I can't remember which way round it goes). We can measure the flux at Earth and look at how it varies over time. We can then deduce the luminosity. If we know the luminosity and (mean I guess) flux then its just simple geometry to work out the distance. This can then be plotted against redshift and Robert's your father's brother.

 

[Dancing David]

Yeah, I was just yanking on Jerome's chain.

And as DRD has pointed out there are other ways that converge on the ratio as well. The Cephids are very cool, mass is related to period and luminosity if I remember correctly.

 

[DeiRenDopa]

Set SIX: 0.267251, 0.725625, 0.934802, 1.08559, 1.19983, 1.29071, 1.45628, 1.69317, 1.78238, 2.03665, 2.10012.

Here are the 'BAC probabilities' I calculate:

"Amaik peaks": 0.00020

"Karlsson peaks": 0.0031

"regular peaks": 0.00011

"DRDS peaks": 0.00014

Set SEVEN: 0.071838, 0.198392, 0.487895, 0.718565, 0.962793, 1.10984, 1.2344, 1.37592, 1.39097, 1.45664, 1.47775, 1.67076, 1.78836, 1.81664, 1.85583, 1.93462, 2.13762, 2.50297, 4.43636.

Here are the 'BAC probabilities' I calculate (note that I have dropped the last object, the one with a redshift of 4.43636, from the calculation, per BAC):

"Amaik peaks": 1.8x10^-5
"Karlsson peaks": 1.1x10^-9
"regular peaks": 1.8x10^-8
"DRDS peaks": 5.5x10^-10
Out of curiosity, I'll try calculating the 'BAC probabilities' for some of 'Arp et al.' cases BeAChooser introduced in this thread, both as per data in the papers and with contemporary data on quasars/QDOs within 30' of the bright, low-z spiral galaxies.

 

[JEROME DA GNOME]

Sophistry without answering a direct question.

Evidence of obfuscation noted.

 

[Dancing David]

Sophistry without answering a direct question.

Evidence of obfuscation noted.

Lack of defense of a sampling bias in statictics, lack of critical argument, lack of anything in terms of critical thought. Please continue to show that you can't defend Arp's use of statistics and that all you can do is name call.

The is yours Jerome.

x 10²³, a whole mole of gnomes.

This is weak argumentation at its worst Jerome, there are pages after pages of critical discussion and you can't manage a little finger of critical thought.

Why not try reading the first five pages and offering a critical defense of Arp's use of statistics?

Can't or won't.

 

[DeiRenDopa]

My attempt at doing 'the BAC calculations' with the 'quasars of NGC 5985'.

Recall BAC's post#329, on NGC 5985 (note that the calculation he details there is wrong):

In this case, observed z = 0.69, 0.81, 1.90, 1.97, 2.13 according to [source] ... Of the above 5 quasars, 4 are aligned.

He didn't say which wasn't aligned, but it's the one with z = 1.90.

Here are the 'BAC probabilities' I calculate for these four:

"Amaik peaks": 0.13

"Karlsson peaks": 0.0047

"regular peaks": 0.0096

"DRDS peaks": 0.068

Note that I used n_k = 6, not 7.

NED gives a whole lot of other "QSOs" within 60' of NGC 5985; in addition to the 5 in BAC's post, and omitting likely duplicates, the rest have redshifts as follows: 0.159846, 0.308804, 0.44584, 0.853491, 1.51803, 1.53915, 1.71782, 1.76805, 1.78098, 1.968, 2.132, 2.18131, 2.53132, 2.59017, 3.059, 3.878.

The following are 'predominantly aligned' with NGC 5985's major axis: 1.51803, 1.53915, 2.59017, and 3.059.

Here are the 'BAC probabilities' I calculate for these four:

"Amaik peaks": 0.023

"Karlsson peaks": 0.075

"regular peaks": 0.0031

"DRDS peaks": 0.017

Note that I used n_k = 6, not 7.

Next, a look at NGC 2639's quasars.

 

[DeiRenDopa]

My attempt at doing 'the BAC calculations' with the 'quasars of NGC 2639'.

Recall BAC's post#329, on NGC 2639 (note that the calculation he details there is wrong):

In this case, [source] identifies observed quasar z = 0.305, 0.323, 0.337, 0.352, 1.304, 2.63 [...] Of the above 6 quasars, 5 are aligned. The paper also mentions some 3 other x-ray sources lying along the axis but I shall ignore them.

Here things are a bit tricky.

First, like BAC, it is wise to ignore the x-ray sources - you need measured redshifts to do 'the BAC calculations'.

Second, the object with a z of 0.337 is listed in NED as a galaxy, not a QSO.

Third, NED seems to have no object, within 60' of NGC 2639, with a z of 2.63.

Fourth, the 'alignment' seems rather odd - of the 4 remaining quasars, only one is predominantly along the minor axis, and one of the other 3 isn't really 'predominantly along' the major axis!

Nevertheless, here are the 'BAC probabilities' I calculate for objects with these five z's (0.305, 0.323, 0.352, 1.304, 2.63):

"Amaik peaks": 0.037

"Karlsson peaks": 6.5x10^-6
"regular peaks": 0 (yep, a probability of zero!)

"DRDS peaks": 0.0040

Note that I used n_k = 6, not 7.

Now NED lists 33 "QSOs" within 60' of NGC 2639; of these, seven are 'predominantly along' the minor axis; they have redshifts of 0.219364, 0.354017 (this is the 0.352 one in BAC's list), 0.9392, 1.12783, 1.55048, 1.57886, and 2.

Here are the 'BAC probabilities' I calculate for these seven objects:

"Amaik peaks": 0.00018

"Karlsson peaks": 0.00024

"regular peaks": 1.3x10^-5
"DRDS peaks": 2.8x10^-5
Note that this time I used n_k = 7.

There are also 7 QSOs listed in NED as being predominantly along the major axis; they have redshifts of 0.305077, 1.30402 (these two are in BAC's list), 1.54253, 1.90763, 2.03032, 2.07549, and 2.89004.

Here are the 'BAC probabilities' I calculate for these seven objects:

"Amaik peaks": 0.074

"Karlsson peaks": 0.00012

"regular peaks": 0.0017

"DRDS peaks": 0.0022

Note that this time I again used n_k = 7.

Next, a comment on NGC 1068.

 

[DeiRenDopa]

Re NGC 1068.

Recall BAC's post#329, on NGC 1068:

Recall that [source1] and [source2] and [source3] collectively list 12 quasars with z =

• [...] Since I don't really know the alignment of these quasars relative to the minor axis, ...

• Oh what a difference of a few years of astronomy surveys make!
  
  First, I could match only 10 of the 12 to what's in NED, within 60' of NGC 1068; a read of Burbidge's paper gave me a tentative match to one more.
  
  So, I could have done what BAC couldn't, get the alignments.
  
  But, it doesn't really matter much ... NED lists 178 QSOs within 60' of NGC 1068! Perhaps as many as 10 are duplicates or possible mis-identifications (there's a quality flag field that's sometimes populated).
  
  And where are these quasars? Aligned 'predominantly along' the minor axis? Or perhaps along the major axis? Nah ... they seem to be distributed entirely randomly throughout the field (if anyone is interested, I could give you some counts, by bins of position angle, if you'd like).
  
  So I'll leave NGC 1068 there for now.
  
  Similarly I'll leave NGC 3516 alone, for now.
  
  Next, what can we conclude from these calculations? Assuming, of course, that I've done them right ... but perhaps I'll never know, BAC seems to have deserted this thread, as has Wrangler ...

 

[Wangler]

Next, what can we conclude from these calculations? Assuming, of course, that I've done them right ... but perhaps I'll never know, BAC seems to have deserted this thread, as has Wrangler ...

I feel bad that I haven't completed my tasks here.

After all, I have been posting in other threads.

I will attempt to at least get my initial calculations done.

I hope we can bring a conclusion to this thread via further discussion, though.

It seems to me that the 'probabilities' that have been calculated are not statistically significant.

In other words, they can't be used to demonstrate some relationship between these QSO's and these galaxies.

 

[Reality Check]

It seems to me that the 'probabilities' that have been calculated are not statistically significant.

In other words, they can't be used to demonstrate some relationship between these QSO's and these galaxies.

I think that is the real point about the use (misuse?) of statistics in the various papers and by BEC in this thread:
How can we tell whether a probability is significant without something to compare it to?


As an example:


I have a die with a million sides each with a different number. It may be weighted toward a set of numbers. Let us test it using BEC's methodology (based on Halton Arp, et al):

1. Throw the die a number of times.
2. Note that for each throw the probability of that number is low (a million to 1).
3. Conclude that the die is weighted to that set of numbers because the probabilities are low.

The proper methodology is to throw the die enough times to get a statistically significant sample and compare the statistics to what we would expect from a unweighted die.

 

[BeAChooser]

I have a die with a million sides each with a different number. It may be weighted toward a set of numbers. Let us test it using BEC's methodology (based on Halton Arp, et al):
1 Throw the die a number of times.
2 Note that for each throw the probability of that number is low (a million to 1).
3 Conclude that the die is weighted to that set of numbers because the probabilities are low.
The proper methodology is to throw the die enough times to get a statistically significant sample and compare the statistics to what we would expect from a unweighted die.

RC, it amazes me that after as many posts as I made describing EXACTLY what I was doing, you folks continue to totally misrepresent the methodology and what it means. That's why I've decided to stop wasting my time with you folks.

A better analogy would be to say that there are million dice scattered about with one number facing up. Each die has a million sides each with a different number.

Now your side believes those dice give no preference to any given number on the dice. On the other hand, Karlsson, Arp and their associates suggest that the dice are loaded and give preference to certain sides of the dice.

Now suppose you randomly pick a group of dice from that field of dice and observe the number that is facing up on each die. You compute the likelihood of that set of dice having numbers as close are those are to Karlson's turning up if the dice were not weighted towards specific sides.

Then you multiply that likelihood by the ratio of the total number of dice in the field over that sample to get a final total likelihood, assuming one could look at the entire population of dice.

And then you ask yourself if you feel comfortable getting that specific total likelihood. If the probability is very small, you shouldn't. If it is small, you must ask yourself whether you are extraordinarily lucky or could there perhaps be some justification to the suggestion that the dice are loaded?

And you then sample the field of dice repeatedly computing an expected likelihood in each case. And as the number of cases climbs where you must assume you were VERY lucky to get the result you got, your confidence that the dice aren't loaded should fall further ... if you are being rational.

There is nothing difficult about this logic, RC. I don't understand why much trouble understanding and restating it.