Yes it is a valid point. However we know the difference between the galaxy redshift and the quasar redshift and that is the expected internal redshift of the quasar. So we can introduce an extra dimension to out plot being that internal redshift, splitting into bands and see if that makes the plot more meaningful. This component can be included.
My red emphasis. Yes, it could, but if it was found that there was indeed a means in nature to produce internal redshifts then it undermines the very reason d'etre for the big bang. It must shake the foundations and put alternative cosmology in a much stronger position. Of course humans are creatures of habit, and that changes only slowly.

[quote=Nereid;1195132]Would you please state whether you are prepared to answer questions on this, and address challenges to it?[quote]
Yes. Yes. Just got rather more replies than I am used to, so might take a little while to catch up.
Certainly I like the idea of making changes so that it becomes a valid test. I think the idea of a control (by shuffling the quasar-galaxy pairs) removes most of the selection effect problems. The selection must be so as to suit Arp's arguments. If he is right then the scatter must be less that the present quasar redshift versus brightness scatter (possibly incorporating internal redshift as a way of dividing the sample up further). If the result is the same with the randomized pairs then Arp is proved wrong, provided the sample is selected in a way that includes quasars and galaxies that arec at about the same distance in the Arp model.
I don't say they are off the table. I say don't expect me to answer them. Look elsewhere.
Whatever the selection criterion used (within a very wide range of possible selections, but excluding hand picking individual objects) the Big Bang people will maintain that all nearby galaxy-quasar pairs with very different redshifts are simply line of sight objects. There is no relationship between the properties of one and the other. It makes no difference in that case if you shuffle the pairs. If it does make a big difference then that disproves the case that they are just line of sight coincidences. Because they are at very different redshifts as well, then it proves that at least one of the redshifts is not cosmological. That the bulk of a redshift is the foundation of the big bang theory.

Yes, that is very helpful, thanks.

I think that my additional idea to compare to the results with the same data with galaxy-quasar pairs shuffled will remove many of the statistical quirks, but it would be good to look at this aspect and confirm that.

Thanks. I do not expect to ever become expert in this matter. But I am happy to be guided by others in it.
I think that some of his papers give certain criterion for selection. If trying to test whether he is right or not, one should logically follow similar criterion. Actually I have sent him an email about this idea to try and get his thoughts about it.
I don't need a million quasars to test this. A thousand will be plenty. The result will be really clear with that many (I tell you this as a statistician).
Well, I think science is best served by having an open mind and investigating all avenues. Yes, of course everyone has different priorities and limited funds. But while a significant number of reasonably distinguished astronomers hold a contrary view it makes sense to test that in an effective way.
Good!

Yes, that is a fair request.

I will try to avoid the limits being probed by present astronomy. Sample size does not need to be a million quasars. A sensible result can be obtained with about 1000 quasars so they should be selected on being relatively bright and possibly relatively lower redshift. This could be done with a cut at an angle in the redshift versus brightness diagram, keeping only the lower left portion. In Arp's view quasars are associated with much brighter galaxies at much lower redshift, so the galaxy survey can be limited to quite a low redshift. I am thinking of the order of 0.1 approximately. I would also exclude all non-spiral galaxies and other galaxies of low absolute magnitude. These criterion need to be refined to fit as nearly as possible with Arp's studies and conclusions. We then have two samples on the same area of sky. I am assuming in what follows that the galaxy sample will be greater than the quasar sample although that might not be the case (otherwise reverse some of the arguments).

Statistically we can then calculate the probability of a galaxy being within a certain radial distance of a quasar. That distance wants to be set to an amount to be at least 100 KPc at the typical distance of the galaxies as that is the distance within which real associations are likely to be. It could be a constant angular separation of a variable one taking account of the galaxy redshift. If there is a galaxy within that distance of the quasar then the closest galaxy is taken as being its "pair" so that a list of pairs results. If the redshift of a galaxy and a quasar are within say delta z of .01 then they will be excluded from the pairs as being real associations.
(question is .01 enough?)

The Test

A scatter diagram is then made with galaxy redshift versus quasar brightness. A trend line is calculated and a range selected about that trend line that includes any concentration of objects that is found. That range is a degree of freedom in the test and selected to give the best result for Arp. The number of objects within and without the range lines is added up.

If it is agreed that in the big bang model a difference of z=.01 is sufficient to say that they are not really associated then according to the big bang all of these pairs are just line of sight coincidences. In that case randomizing the pairings should make no difference to the statistics whatsoever

As many as desired random pairings can be done, but say 100 to start with. In each case the same trend line and range about that are then used to see how many pairs fall within and without the range. The true pairings are then tested against the 100 cases to see if they fall inside or outside the range of them. If inside then we reject the Arp hypothesis that quasars are are associated with galaxies at lower redshifts. If they fall outside then at a 1% level we reject the hypothesis that the pairing are random. That means they are true associations even though at different redshifts and that at least one of the redshifts (presumably quasars) is not a reliable measure of distance.

There is a way to test for higher levels of significance without doing millions of randomized runs, but I can explain that another time if we get well outside the 1% (as I expect that we will).

Good. That definition was likely used in most of Arp's papers, whether he was aware of it or not. Can you think of any reasons this might make your work difficult?

Actually, you could. That's rather the point. My "haphazard" comment was with regards to how they were selected: there weren't very many systematic surveys for quasars at the time, because people didn't have a good idea of what exactly to look for. Many were found because they are strong radio sources and identified during early radio surveys. But it turns out that maybe only 10% of quasars are radio loud. There's a big bias right there. If a bunch of galaxies were observed in radio, it might cause more quasars to be identified around galaxies than in the field.

Do you understand the nature of this selection effect? Do you understand how it could produce a plot like the one you showed above?

I must say, that I'm quite happy to see that you've worked out some of the numbers here: that's certainly more than many ATM threads! I don't have time to check your math, except for your redshift: z~=0.008 corresponds to 30MPc, not 0.04 (though your angular size is about right for that distance). So you'll have to check the z=0.2 angle as well...

Keep in mind, though, that SDSS has hundreds of thousands of galaxies out to z~=0.1, so there will be lots of quasar/galaxy pairs (and multiples).

There are a few other reasons that quasars might be slightly more numerous around galaxies: lensing, both strong and weak, is an example, as Nereid mentioned. Both of those effects have been investigated and published in the literature.

Also, what do you have to say about the various galaxies that have been identified out to redshifts of >2? They are not quasars, by any definition of the term, but do have redshifts in the same range as the quasars that you are claiming have intrinsic redshift components.

And you definitely need to investigate the use of mock catalogs. They are very important for any kind of statistical work, and cannot be completely replaced by your shuffling method.

Unfortunately, that's not necessarily true. A thousand biased quasars won't do you any good at all, and is in fact exactly the problem encountered by other intrinsic redshift proponents.

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

I fail to see why that kind of attitude is needed.

__________________
"Stupidity gets denser in a crowd" - Old Finnish saying. [My website and My BLOG] [Nimblebrain forums]

No. Especially with the shuffling comparison.
OK, if Arp's studies were done mostly with radio loud quasars then that should certainly be considered as a separate test here.
No. I cannot see why for galaxies with a larger redshift all the quasars near then should suddenly be closer if they have no real association.
Yes, I made a slip there, I now get z~.007 so must have used a slightly different value for H than you.
By selecting only the nearest to each quasar that will reduce the sample as long as not too many quasars are used.
OK, I may be displaying some ignorance here, but from what I know about optics (and lensing is essentially optics) a convex lens will always make background objects further away. To the extent that a lens (that drops off in effect beyond some distance) brightens some background objects it will dim others. Is that not so?

At any rate this is not a problem except where a quasar is multiply lensed , so such cases may need to have one case excluded. The brightness effects of lenses will tend to obscure the true relationship of brightness with distance and to that extent make it slightly harder for the Arp case to succeed, but I don't think it is significant enough to make a big difference. Unless a big proportion of quasars are considered to be greatly affected.
These galaxies are accepted as having reliable redshifts because the brightness versus redshift scatter diagram is (as one would expect) showing a solid correlation. Generally a big proportion of quasar redshifts are expected to have an internal redshift and only a very small proportion of galaxies, and those mostly active galaxies.
Why is that? I am not disagreeing, just seeking clarification.
Well a million biased ones won't help either. So the bias needs to be contained.

This is a question I wanted to raise too. How does this affect this claim:
since we now observe galaxies in the redshift range of quasars?

Well I am trying to test the Arp view versus the Big Bang view. AFAIK Arp accepts that galaxy redshifts are generally reliable. There are a few cases where galaxies have connecting bridges and somewhat different redshifts, but the differences in redshift for galaxies and quasars that appear connected are far greater. In the cases that I have seen at least one of the galaxies involved has been active, which I take to mean it is somewhere between a quasar and a galaxy in its evolution. I don't know if that is the standard view.

I repeat that the logical test for this is whether the objects make a sensible scatter diagram of redshift versus brightness. If these new galaxies are sufficiently dim then they will continue the reasonably tight relationship of previously studied galaxies. However quasars have hardly any relationship between their redshift and brightness, which is why the redshift is suspicious in the first place. That is what this test hopes to exploit, because it gives a possible means of showing a relationship between redshift (of a galaxy possibly really associated with the quasar) and the quasar's brightness. If this succeeds then it immediately shows that quasars do not have such a huge range of brightness as previously thought. The situation is complicated by the fact that even in Arp's view, quasars go through a life cycle that involves rapid increase in brightness as the internal redshift fades away. However we can test for this very accurately with this method because we know the difference between the quasar and galaxy redshifts and that is the quasar internal redshift (except if it is a genuine line of sight object).

So the test may realistically need a 3D plot of quasar brightness as a function of galaxy redshift and implied internal redshift (i.e quasar redshift less galaxy redshift). Or plotting it in slices of internal redshift range.

Please reread my comments above about Bell's paper. Grabbing a catalog (which is clearly marked as not to be used for statistical analyses) and making plots with it, without understanding what the data actually represents is not going to earn one any respect. Doing so repeatedly for many years is a way to become ignored, at best.

Incase my points above aren't clear, here's a similar plot to the one given by Bell (boy did I have trouble finding the right catalog to do this with...). It has V magnitude on the x-axis and a measure of distance on the y-axis (admittedly, inverted). What conclusions should be drawn from it?

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

Any respectable astophysicist can easily fit this distribution with a straight line.

__________________

Any comments in glorious red are to be considered in ModeratorMode.


善數, 不用籌策 (shàn shù, bù yòng chóu cè)
He who is good at counting, uses no counting tools
“A good scientist has freed himself of concepts and keeps his mind open to what is”
道德經, 二十七 (dào dé jīng, 27)

You're not quite right: since it is already a log/log plot, that'd be a power law! Or, maybe a broken power law. Can't you see the knee there?

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

Ah, I stand corrected!!!!

__________________

Any comments in glorious red are to be considered in ModeratorMode.


善數, 不用籌策 (shàn shù, bù yòng chóu cè)
He who is good at counting, uses no counting tools
“A good scientist has freed himself of concepts and keeps his mind open to what is”
道德經, 二十七 (dào dé jīng, 27)

What I see is essentially a triangle. It all makes sense except that the right hand side of the triangle is not vertical. I do not understand why there are not many more objects with magnitude 10-12 and very small parallax. Has there been some selection effect, such as including dim stars that are known to be near to us?

This is a good example for the exercise at hand. Thanks for it.

The plot should contain all the stars from the Hipparcos main catalog that have reliably measured parallaxes, which is essentially complete to V~12.1, I think. Remember, that's apparent magnitude, just like the plot in Bell's paper.

Edit: wait... No, it isn't complete to V~12.1, which is part of the problem...

Try and come up with an explanation for all the main features of the plot. It is a good exercise. Stars are relatively easy to understand, compared with quasars.

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

Leaving aside - for now - what might constitute "relatively bright" and "relatively lower redshift", please explain why "[a] sensible result can be obtained with about 1000 quasars".

Also, what constraints (if any) are you proposing be placed on the number of galaxies in the sample? Please explain your answer.
It might be helpful to refer to a dummy/mock diagram, not least so that readers can identify what "the lower left portion" means!

Also, what does "done with a cut at an angle" mean?

For example, is there a range within which the angle should fall?

And what criteria do you propose be used to determine the angle to be used?
Why introduce Arp?

The proposal in the OP is quite explicit*; a disproof of the cosmological nature of (all) quasar redshifts.

Have you modified the ATM proposal? If so, would you please re-state it, in a form that you are prepared to answer questions on, and address challenges to?

While I'm at it, could you please clarify this: do you propose to "prove" that the observed redshifts of all quasars are non-cosmological?
Why?

What (objective) criteria do you propose to use to determine the z cut-off?
Why?

What criteria do you propose to determine the 'absolute magnitude' cut-off?Again, why?

The proposal in the OP does not mention Arp, nor does it seem to imply that it is a test of any of Arp's conclusions.

Surely the criteria "need to be refined to fit as nearly as possible with [...] studies and conclusions" consistent with concordance cosmological models, since that's what the proposal in the OP explicitly seeks to overthrow?What criteria do you propose to select the area of sky?How? In detail, please.What methods do you propose to estimate the distances to each of the galaxies in the sample? Or, perhaps, what sources do you propose to use to obtain those distances?

What is the rationale for using "at least 100 KPc"?Please describe in more detail how you propose to do this, using dummy/mock values.How do you propose "quasar brightness" will be determined?Details please; for example, how do you propose to identify a "concentration of objects"?How do you propose to determine this cut-off?Details please; for example, what method of randomizing? what statistics?How do you propose to determine the optimal number of random pairings?Details please.

Specifically, how does this proposed method address the known physics/astronomy of quasar variability, reddening, and gravitational lensing (both strong and weak), to give just a few potentially confounding factors?I have many, many other questions about the proposed test, but they will have to wait for at least some answers to the questions here.

... except for "quasars [...] should be selected on being relatively bright".

What are your current thoughts on how the quasars should be selected?

I appreciate that this question will, very likely, be one of the toughest to address.

However, as the proposal in the OP explicitly concerns "the big bang"^, I expect that you will offer a detailed rationale for selecting quasars "bright" in the x-ray AND radio bands ...

* "This proposal is that there exists demonstrable real associations between galaxies and quasars that are at very different redshifts. That alone disproves the cosmological nature of quasar redshifts and totally undermines all the interpretations of the big bang."

^ I think you mean "the big bang theory"!

Yes, that was my conclusion that it couldn't be complete to mag 12 but only to about mag 9. It looks to me like known closer stars (maybe based on spectral type or HR position or something like that) have been included to mag 12.

The lower left slope is correct for the dimming of stars as 1/distance^2. The horizontal top just means that there is a limit to closeness of all stars, but I am surprised that there are not more fainter stars than bright ones at the same parallax. (near the top)

I worked out that there should be a few more stars to the left at the vertex, and then I realized that there are - Sirius and Canopus are missing!

(emphasis added).

I'm confused ... the proposal, as currently presented, would not seem to have anything but the most marginal relevance to "all the quasars near the[m]". Can you clarify please?
IIRC, this issue came up in one of the other, earlier, ATM threads that you started, and it was resolved to your satisfaction.

Why not start a thread in BAUT's Q&A section, on this, so you can get as deep an understanding of the relevant physics (and astronomy) as you need to address this in the design of your proposed test?

If you then still propose "[a]t any rate this is not a problem except where a quasar is multiply lensed", please be sure to state this explicitly ... and give BAUT members the opportunity to question, and challenge, whatever ATM idea(s) you propose to incorporate in the test ...

Nope. Neither is it complete to V~9. Keep trying. It's a bit more complicated than you might think.

Why? Give me a bit more detail there... Also, Ari, this is all for your benefit as well.

There's a reason for that. Remember, the magnitudes are apparent, not absolute.

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

It is irrelevant who's right and who's wrong, I was just wondering why is there a need to call any serious work (no matter how wrong it actually is) a comedy routine? Isn't that insulting? And that part of discussion was generally about studying intrinsic redshift issues, so you called works of everyone doing that "a comedy routine". I found that specifically strange comment because dgruss23 is one of those people who has published on intrinsic redshift issue. So, in my view, your comment was very unfortunate, at best. If you think people are doing so bad work that they should be ignored, then go ahead and ignore them, but there's really no need to start throwing insults at them at any case.

__________________
"Stupidity gets denser in a crowd" - Old Finnish saying. [My website and My BLOG] [Nimblebrain forums]

It seems to me that there's an unjustified assumption here, Ray. This assumes that quasars have some standard brightness in this alternative scenario. I don't think that's necessarily good assumption. Think about Arp's model, which you have yourself brought up here. I think there's a good possibility that the ejected objects from galaxies (that will be quasars according to Arp's model) are originally of different sizes, and therefore have different apparent brightness. That means that, in addition to dgruss23's comment about brightness evolution, you potentially have a random brightness distribution to start with, meaning that your test doesn't prove anything if it shows random distribution for brightness.

__________________
"Stupidity gets denser in a crowd" - Old Finnish saying. [My website and My BLOG] [Nimblebrain forums]

That it'd make a nice Valentine's Day present?

I think that I must have been sleepy when I did that post, I can't understand it myself.

The problem is with the word closer being mixed up with meaning nearer to us and at lower angular separation I think. Let me try again.

I can see no reason why gravitational lensing would make more quasars appear at lower angular separation from a galaxy, except for the case of multiple lenses which I agree would need to be excluded (if that is possible ... I suppose by looking for nearly identical redshifts).

Yes, we may have been there before.

Yes, but when they have the same parallax then that amounts to the same thing, so my comment about the top part is correct.

I feel like I am being tested. Can't see the point.

Fair comment. I did add elsewhere that even in Arp's model quasar brightness changes a lot. Therefore it makes sense to analyze quasars in groups by implied intrinsic redshift (being quasar redshift less galaxy redshift) and see if each one has proper behaviour. It needs to be checked that this doesn't somehow beg the question.

However the main point is that the present scatter diagram is almost unrelated fro quasars, so even if they have quite a large variation in brightness they could still get a tighter relationship thanthe present one.

I think I will have a further thought on this and maybe make some test data to see how much of a problem this is.

It isn't necessarily my opinion, but rather the view I've seen whenever anything "Arpian" is mentioned in astronomy circles (which doesn't happen very often). I've heard people wonder why an intrinsic redshift paper wasn't published on April First (a day of foolishness in the US, Ari: I don't know if it is similar in Scandinavia). Perhaps it is insulting, but the quality of the papers is generally laughable.

If you (or dgruss23, or someone else) is aware of an intrinsic/quantized redshift paper that actually pays attention to the issues I've listed above, please let me know. I certainly haven't done a literature search, but neither the ones I've seen listed on BAUT, nor those I've seen posted to astro-ph recently have anything worth paying attention to. That is my opinion (that and a dollar won't buy you a cup'a in most places these days...), but I think it is pretty common (among those who actually bother to consider them at all).

And speaking of opinions, I mostly just feel sad that Arp, Hoyle and the Burbidges (all once well respected astronomers) have essentially gone off the loony end. Arp is still fighting a battle he lost 30 years ago. I don't know anything about Bell's or Mcdiarmid's earlier work, so I don't know quite how I feel about them.

I've heard that dgruss23 has published, but I've not seen the papers, so have no opinion about them yet. Linky, linky?

As I said, intrinsic/quantized redshift is ignored, at best, among astronomers.

__________________
"What do you care what other people think?" -- Richard Feynman
"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Feynman, at the conclusion of his Challenger report

snipping the 'sleepy' bits ...Perhaps now would be a good time then to review this topic?

Once you do, I think you'll appreciate that, under 'the big bang theory', and with the proposed test as it is currently drafted, a non-null finding (that the number of galaxy-quasar pairs/associations is greater than what would be expected if galaxy-quasar pairing/association were purely random ... or something like this) is to be expected ... but that this pairing/association is due to gravitational lensing (both weak and strong), not physical (3D space) proximity.

IIRC, parejkoj (or was it matt.o? dgruss23??) already alluded to this, in an earlier post in this thread, pointing out that there are a number of papers in the literature on the consistency between the expected signal and the cold, hard observational 'ground truth'.

So, when do you expect to have the next draft of the proposed test ready for BAUT readers to review?

Sigh... as you don't have enough common courtesy to leave that kind of things unsaid, I don't want to continue this discussion further.

__________________
"Stupidity gets denser in a crowd" - Old Finnish saying. [My website and My BLOG] [Nimblebrain forums]

Here's a linky.
http://arxiv.org/ftp/astro-ph/papers/0408/0408348.pdf