Evidence of Gravitational Waves, or Confirmation Bias?

[Zyxzevn]

You crack me up.

This is a personal attack.
Want to know what cracks me up about you?

[Zyxzevn]

This bias is a problem that is common in science.
The statistics are used to cover up the bad science behind it.

First they model the unknown and define how this unknown behaves.
In LIGO this unknown is the noise. It is really gigantic.

Their claim is that they know the unknown by identifying the noise that is
very common. And they claim that all unknown has been identified by
these common examples.
This is where I protest. You have to identify every unknown, also the uncommon ones.

They also have a signal signature of a new thing that they want to be true.
And as we can see in their produced filtered signal of unknown origin,
that this signature partially matches this signal.
So their conclusion is that this unknown signal must be their modelled signal.

And I protest here again. The match is not good enough.
A part of the signature is not in the filtered signal. This is an essential problem.
And a large part of the signal is not visible due to too much noise.

Then the LIGO scientists fiddle the statistics in such a way that they claim it was
a very rare incident (high sigma).
This is mainly based on:

1) the shape of the signature
My protest:
a) Noise of electric origin can easily cause the same signature.
b) there are systems within and near the LIGO that can cause similar signatures.
c) The filter can cause shapes similar to the signature.
d) According to one LIGO researcher, signatures were found very often.

2) the frequency range of the signature
My protest:
a) electricity is in all ranges. 60Hz is common, other ranges are just rarer.
b) within the frequency range there are resonating systems within the LIGO
(like their own laser-mirror system).
c) They filtered away so much you can't see what signatures are from non-linear noise signals.

3) the occurrence at 2 places.
My protest:
a) We do not know how often these signatures appear at one place.
b) There might be a common atmospheric signal or plasma from the sun.
c) There might be a common signal caused by earthquakes (and piezoelectricity?)
d) Communication is possible through the electric grid

Therefore my proposal:
exactly model the unknowns. Model their signals perfectly, so you
can subtract them from the measured signal.
Note: this is different from filtering their common frequencies out, which
hides a part of the noise and makes noise as cause of the signal less visible.

Otherwise you are just creating a new face on Mars.

In short: You can only identify a new thing by first identifying all unknowns.

[Michael Mozina]

This bias is a problem that is common in science.
The statistics are used to cover up the bad science behind it.

That's certainly true as it relates to recent LIGO claims. Not only does the sigma figure they provided *not* eliminate other environmental causes of the chirp signals, they don't even offer us a quantified figure to calculate the odds that it must be celestial in origin vs. environmental in origin.

The whole "process of elimination" methodology that was applied to other environmental influences was simply tossed out the window in the case of all celestial origin claims. Not only is the sigma figure irrelevant, it was *entirely based on cherry picked data*. Sheesh. Pure "bad science".

[Higgsy]

The fact they're already 0 for 3 demonstrates that they don't know what they're talking about.

So you think the LIGO collaboration doesn't know what it's talking about but you do. Good luck with that.

[Higgsy]

You crack me up.

This is a personal attack.
Want to know what cracks me up about you?

Saying that I find what you say is funny is a personal attack? The idea that you are doing a thorough job of the data analysis while claiming the LIGO team doesn't know how to analyse the data is very funny. It really is.

[Higgsy]

This bias is a problem that is common in science.
The statistics are used to cover up the bad science behind it.

SNIP

In short: You can only identify a new thing by first identifying all unknowns.

Have you worked through that tutorial yet?

[Michael Mozina]

The fact they're already 0 for 3 demonstrates that they don't know what they're talking about.

So you think the LIGO collaboration doesn't know what it's talking about but you do. Good luck with that.

In terms of their reliance upon improper and highly biased methodology, nope, I don't think they have a clue what they're doing. If they followed their own process of elimination methodology consistently, there would be a category in their methodology for "unknown cause" and I wouldn't already be three for three in terms of successfully predicting their utter failure in being able to support these claims visually.

There's not even a logical way to falsify their claims, and nobody outside of LIGO could ever hope to replicate or verify their claims either. It's a closed circular feedback loop in terms of their claim, and LIGO won't even answer even the most rudimentary questions about the specific veto that took place of this exact signal. They even hid the specifics of that veto from the peer reviewers.

What would you even accept as a falsification of their claims? Unlike LIGO, I am able to offer you a clear and precise method by which they can falsify my claims and change my opinions. What happens once the next LIGO detector comes online if these very same 'chirps' are sometimes observed in just two of the LIGO detectors, but not all three? What happens if they get to 10 such claims and not a single one of them can be confirmed visually, even with better "triangulation methods"? Is there anything which you might accept as a falsification mechanism?

This entire "discovery"" claim is based on hypothetical unseen waves from hypothetical unseen objects. There's no logical way to even verify them or falsify their claims in an independent manner without spending *billions* of dollars building at least 3 LIGO detector. Even then nobody could falsify any claims that LIGO might make between now and then. The whole methodology reeks of "bad science" and confirmation bias which is exactly why they are already 0 for 3 in offering any type of external support of their claims.

Wanna make a bet that I go 4 for 4 in my prediction that they cannot verify their claim externally, and LIGO goes 0 for 4 in their next paper?

LIGO hasn't even made any useful or meaningful predictions that wouldn't necessarily apply to a *NULL HYPOTHESIS*! Even a NULL hypothesis would *predict* that LIGO can't and won't be able to verify their claim visually or via neutrinos. What a pitiful theory if they cannot even make any useful predictions that are any different from a NULL hypothesis, and there's no way to *falsify* the claims?

Get real. There's simply no rational excuse for their blatant confirmation bias problem.

[Zyxzevn]

Have you worked through that tutorial yet?

Of course..
Did you?
Of course not.
You can not program.

[Zyxzevn]

You crack me up.

This is a personal attack.
Want to know what cracks me up about you?

Saying that I find what you say is funny is a personal attack? The idea that you are doing a thorough job of the data analysis while claiming the LIGO team doesn't know how to analyse the data is very funny. It really is.

It is still a personal attack.
I do claim that the LIGO team has omitted a part of the analysis,
as I have explained above many times.
But your brain does not seem to capture that, because you are too biased.

Instead you could help me and the LIGO team by point me out
the missing paper that deals with the resonances and the dynamics in the system.
It is their job, not mine.

[Higgsy]

Have you worked through that tutorial yet?

Of course..

No you haven't. Don't tell porkies. If you had, you would know and have acknowledged that there are no tones in the raw data at 134Hz or 268Hz. Your idea that there should be content in the raw data at those frequencies is simply belied by the data itself. Which you would know if you had worked through the tutorial. Which you obviously haven't.

To prove me wrong all you have to do is to show how the raw data contains those frequencies, by starting with the raw data and showing step by step the processes from which it can be unequivocally shown that those tones are there. I know you won't be able to do that because those tones do not exist in the raw data.

If you had worked through the tutorial you would have seen how the signal is revealed from the raw data by whitening and band passing.

Why did you claim to have worked througgh the tutorial when you plainly haven't?

[Higgsy]

I do claim that the LIGO team has omitted a part of the analysis,
as I have explained above many times.
But your brain does not seem to capture that, because you are too biased.

And I have explained many times that the power and signal recycling does and cannot introduce tones into the raw data in the way that you expect, and that your expectation is based on ignorance of interferometry, and that, moreover, the frequencies that you predict based on your flawed understanding are simply not present in the data. What more do you need to throw your silly hypothesis in the bin than the fact that your prediction is simply wrong to fact.

Instead you could help me and the LIGO team by point me out
the missing paper that deals with the resonances and the dynamics in the system.
It is their job, not mine.

There are many papers dealing with how noise is managed in the system. For example LIGO-P1500238-v24, LIGO-L1500138-v6, LIGO-P1500229-v31, arXiv:1411.4547v1, LIGO-P1400177-v5, and arXiv:1604.00439. When you have read, understood and absorbed the information in those papers, come back and we can talk.

If you are referring to your fantasy 134 and 268Hz, please try to understand that those tones are not present.

[Cargo]

The really sad thing is all these people really believe in Black Holes. That's where the problem starts. LIGO can claim anything they want, it won't make it real.

[Michael Mozina]

http://www.internationalskeptics.com/fo ... stcount=62

Selfsim: I think another implication here is that noise transients, such as 'blip' or 'chirp' transients, occurring non-simultaneously (within the c window 'lag') across both detectors, would be also be ruled out as being at play here using this technique. If the 'signatures' of either of these phenomena remained following this process, then this would be evidence of propagation at c. Where no corresponding EM evidence exists in the PEM sensor array logs, one could then perhaps deduce a celestial (GW) cause. If such evidence did appear in the corresponding PEM sensor array logs however, then one couldn't deduce celestial GW type causal mechanisms (eg: a BH-BH merger).

As it turns out, there was no corresponding simultaneous PEM log data, so celestial GW causal mechanisms thus account for what remains in the combined site h(t) data.

I think this couple of paragraphs perfectly illustrates the ease at which someone can make "assumptions" that only favor their own beliefs. I think it's safe to assume that whatever these signals might be, they're "unusual" events that do not occur on a daily basis, but they may occur several times a year. Whatever the actual cause of the noise, it's likely to be a relatively 'rare' event of unknown origin. Now we're trying to ascertain what that source might be.

One might just as well claim that because no celestial events were observed during the timeline in question from the region in question, it's causal mechanism must be EM related.

The bottom line is that we have no evidence to support any specific claim as to the origin of these three signals. While they're rare in frequency, they are reoccurring events. The logical thing to do here would be to double down on the sensitivity of the PEM detectors and other types of equipment, and to analyze what *exactly* set off the offending veto of the first signal. Whatever the LIGO folks were trying to "veto" out of the data by writing that veto in the first place, is probably a good place to start in terms of actually figuring out the real cause of similar such signals.

LIGO however will not tell anyone what that veto was designed to veto out of the data set when it was first written. They won't explain how or why that veto achieved a "high confidence" in terms of it's rejection of the signal. They won't quantify *anything at all* related to the "safety" of any "unsafe vetoes" either. Their entire approach has been to hide this critical information, first from the peer reviewers, and then anyone else from the outside of the organization that has legitimate scientific questions about this veto. They are literally making it *impossible* to either confirm or to falsify their claims independently by withholding *vital* information related to that veto.

If we were being entirely "intellectually honest" with ourselves, we have to admit that we have no identified source of these signals to be found in any external data sets. If we're doing this "by the book" in terms of physics, this signal, and all similar chirp signals should fall into the "unknown origin" category. The moment we try to exempt only one specific possibility as the cause the signal from being eliminated due to external evidence, we automatically bias our claim in the direction of the thing that we excluded from the process of elimination and which we applied to everything else.

We also have to be honest with ourselves and acknowledge that whatever the actual "cause" of blip transient events might be, we do know their origin, and they also tend to fit the same mass merger models. We also have only a highly limited amount of blip transient data to work with, and we really can't be sure of all the different types of signals that might be caused by blip transients.

Since blip transient events fit the same mass merger models, the only way to start to try to differentiate between blip transients and GW events is to start making highly dubious claims about what we might "know" about blip transient events, without even knowing their actually cause. The detectors had just be upgraded by a factor of 10 in terms of distance, and by a factor of 1000 in terms of overlapping volume. It's certainly possible that an order of magnitude improvement in sensitivity might allow both detectors to pick out *rare* EM discharge events at both detectors, whereas the older equipment simply didn't have the necessarily level of sensitivity. Keep in mind that this first correlated signal took place during the *engineering* run that was supposed to start collecting some new data related to these questions.

After watching the video about LIGO, it's really clear that they were "excited" to finally see a signal that was cross correlated in both detectors. That is a necessary ingredient in terms of GW but it doesn't guarantee that the signal *must* be GW related. They were also excited by the strength of the signal compared to ordinary background noise. Not only was this a "rare" event in terms of that correlation, it was also a relatively "strong" event.

What LIGO never was able to do however is A) determine the cause of ordinary blip transients in order to actually rule them out, and B) they steadfastly refuse answer any specific questions about the veto that took place. This tells me that they wrote that veto with a specific intent in mind, and they simply *refuse* to share the intent of that veto with anyone outside of LIGO. They are definitely hiding something important by not releasing this *vital* information.

The sigma calculation that LIGO provided is nothing more than a psychological Trojan horse. It not only does *not* directly support their claim as to cause, it doesn't even rule out ordinary environmental influences because all of those influences were stripped from the cherry picked data *before* the sigma calculation took place. That calculation can't rule out blip transient events either, and blip transients cannot be ruled out without making a *host* of assumptions that simply cannot be supported at this time. It's ultimately a meaningless sigma figure because LIGO simply "cherry picked" the data they *needed* to create a high sigma figure from a larger data set which would *never* have supported their conclusions. That's why they had to cherry pick only very "quiet" times and non vetoed times. That sigma figure is entirely *contrived* with the express intent of "appearing" to eliminate environmental factors, when in fact it simply does not do anything of the sort.

The only way they can try to claim to rule out environmental factors is by applying their process of elimination methodology to *everything else first* and then by blatantly exempting their own claims from any type of process of elimination.

I think the LIGO team has done an excellent job in their analysis, which is being demonstrated by its robustness whilst enduring quite 'gritty' legitimate scientific scrutiny. Great to see the process in action, I must say.

LOL! LIGO is afraid to even answer any legitimate scientific questions related to that all important veto of these chirp type signals. They won't explain why the veto was first written nor explain what kind of noise that it was intended to "veto" out to start with. They won't explain how the veto in question assigned a confidence factor to the veto. They refuse to quantify anything related to "safety" in terms of this veto, or any veto which they deemed 'unsafe', so again, their contrived sigma figure means absolutely nothing with respect to cause.

In terms of pure methodology, the GW claims are probably the single worst example I've ever seen in terms of confirmation bias. There isn't a single part of their methodology that doesn't play into their bias. It began with the "cherry picking" of hand selected data. It continued with hiding critical veto information from the peer reviewers and from the public. It required a host of unsupportable assumptions about the nature of blip transients without even knowing their cause. It required a blatant *confirmation bias* problem which forbids any signal from being labeled "unknown" in origin, and it required an exclusion of their own claims from the process of elimination methodology which LIGO applied to everything else. All five of those flaws that I cited in my paper came into play in this highly biased process, and each and every single one of them favored GW's, and disfavored everything else.

This really is a physics train wreck in slow motion. There might be a way to demonstrate that blip transients can effect multiple detectors without having an effect on all of them when we have more than 2 of them online, but at the moment, their isn't even a way to "test" their claims related to blip transient events, and they can't explain the cause of blip transient events. Once there are three detectors online however, the train wreck is inevitable.

This charade can only last until their are three stations online. Once that happens, there might finally be a way to falsify their claims about blip transient events. Without a third detector however, it's not even possible to do falsify any part of their claim.

Now of course without all those flaws in their *methods*, these signals would not have been categorized as GW in origin in the first place, but there is no way to refute or support their claim *externally*. They've made it completely impossible for anyone outside of LIGO to falsify their claims. Since LIGO won't even discuss the only veto event that might help to determine the real cause of these chirp type signals, the best anyone might hope to do is to try to find some type of flaw in their *methodology*. I've found five of them and there may be more.

That last paper on Arxiv tries to suggest that there might be a sixth flaw in LIGO's methodology, but even if that one item is determined to not be a legitimate flaw, the five problems that I listed are still applicable. Only one of the five primary methodology flaws cited in my paper have even been debated. Nobody seems to dispute that the fuzzy sigma number doesn't directly support LIGO's claims as to cause, and it doesn't rule out any environmental variables either, nor blip transient events. Nobody seems to wish to even discuss the confirmation bias problems.

Had they visually confirmed that a celestial event took place at the same time and at about the same distance that LIGO claimed, I'd actually be happy for them and happy for physics. As it stands however, this has all the earmarks of a scientific train wreck.

It's just absurd to think that we're spending billions of taxpayer dollars on massively sophisticated hardware to test these ideas, while relying upon two bit biased methodology.

[Metryq]

Maybe the signals are "perytons" from the ISS.

[kell1990]

I've followed this thread since its inception, and one of the things that has become abundantly clear is that the--dare I say it--the "standard model"-- is what the researchers now think is the truth of this matter. It is their way or the highway. Just read some of the overbearing rhetoric from some of them and you'll quickly see that this is the case. "higgsy" is a perfect case in point. (probably a newly-minted PhD, and thus knows everything there is to know about his subject...)

This "standard model" includes dark matter, dark energy and black holes.

The problem is that only about 4% of the activity in the universe can be attributed to gravity. Why is that? Because if the other elements aren't included--like dark matter and dark energy--the entire theory falls apart. So the theorists have invented these other elements--which no one has ever seen or measured--to make the theory work.

I say again, in the immortal words of Richard Feynmann, "If the theory doesn't match the experimental data, then the theory is wrong." (I don't think that's an exact quote but it is right on target when it really gets down to it.)

I've read the statements of "higgsy", and I've read some of the stuff that Brian Coberlain has put up, as well as some of the other statements that their allies have pronounced.

Their statements are what Feynmann was talking about. Their statements don't match reality.

And that's what this entire argument is about: Reality.

It is so much simpler to accept the fact that the electromagnetic force influences the behavior of the universe rather than cling to the nonsensical notion that gravity is the predominate force in the universe.

For the life of me, I can't see what is so hard to understand about that.