Evidence of Gravitational Waves, or Confirmation Bias?

[Michael Mozina]

I think these gravitational wave claims clearly demonstrate the danger of getting 'used to' the idea of invisible stuff ruling the universe. Once you get comfortable with 'dark energy" and "dark matter', "space expansion" and "black holes', the sky is literally the limit in terms of what you're likely to accept without question. They're already up to 95 invisible stuff, and they keep adding more invisible stuff by the day.

Now we have hypothetical invisible things called "black holes" being used to "predict" that no detectable EM radiation occurs when invisible things release invisible waves that show up in detectors here on Earth. This isn't science anymore, this is myth-making with math. There's not a single control mechanism to be found in any of this nonsense.

Every single one of the three gravitational wave claims is devoid of any visual support, and every single one suffers from the same confirmation bias problems. They all suffer from every other criticism of LIGO's methodology that is listed in my paper too with the possible exception of the conflicting account of veto events surrounding GW150914.

There's ultimately no correlation between the sigma figures and their claim as to cause. There's no external corroboration for any of these claims. These claims are entirely dependent upon LIGO, and only LIGO data, and LIGO has a proven track record of misrepresenting the historical facts as it relates to vetoes, and engaging in blatant confirmation bias, and bad scientific methodology.

I'm really dismayed at the state of affairs of astronomy today. There's literally no empirical difference between astrology and astronomy today in the lab. They both make wild claims about things which fail to show up, or that can be confirmed in the lab, and they both make grandiose claims about mostly invisible forces controlling the fate of the universe. Empirical physics has become the arch enemy of both astrology and astronomy today. It's just sad.

Wake me up if LIGO *ever* publishes a paper with a visual confirmation of their claim. After that erroneous nonsense in their published paper that there were no vetoes present within an hour of the event, and their avoidance of my questions about the veto which actually occurred, I don't trust them as far as I can throw them. I don't believe any of their nonsense because it's all based on highly biased and ridiculously flawed methodology, and they tend to overtly misrepresent and whitewash historical facts at will.

I openly wondered after the first paper what would happen if they kept *not* finding any visual confirmation of their claims. I guess I have my answer. It's full steam ahead, and logic be damned.

[Michael Mozina]

I must say.....

The mere fact that there wasn't a single gamma ray burst that was detected from even one of these three events sure makes LIGO's claims of celestial origin of these signals look absurd. These events are presumably releasing the energy equivalent of multiple solar masses in about a 1/4 of a second, but somehow, as if by pure magic, they don't ever release enough EM energy to even set off a single GRB detection. How does that work?

I'd be *thrilled* if LIGO had discovered gravitational waves from a couple of nearby massive objects in orbit around each other and created a cyclical signal, which could also be verified visually. I'd have also been thrilled had even one or two of these three "transient" events been verified by *anything* in our massively expensive telescope inventory too. As it stands however, this is starting to look like pure desperation IMO. If one incredible claim doesn't sound convincing enough, just make a few more of them and people will start to believe them eventually.

If I was skeptical in the past based on two events without any sort of visual confirmation, I'm even more skeptical now. The complete lack of a visual confirmation undermines their claim that these types of signals are celestial in origin. If they were celestial in origin, where's the visual beef?

https://www.youtube.com/watch?v=Ug75diEyiA0

[Zyxzevn]

Well, the good news is:
the more LIGO detects, the more certain it is not a gravitational wave.

Again, the real theoretical shape is not visible in their graph.
According to the real GW theory, the waves must gradually grow in frequency and amplitude together.
So even if the GW theory is correct, these signals are exactly not what they predicted.
The only reason that they think that it is a GW, is because they could not find another explanation, yet.
That is because these PhD students do not understand EM and non-linear stochastic signal processing.

[Higgsy]

LIGO "found" a new gravitational wave.. (on January 4, 2017)

GW170104: Observation of a 50-Solar-Mass Binary Black Hole Coalescence at Redshift 0.2

https://journals.aps.org/prl/abstract/1 ... 118.221101

Redshift 0.2 is quite near, I think, and should be visible.

I have read the paper but I need some help with some things. First of all how far is z=0.2, say, in light-years? At whatever that distance is, what is the difference between the apparent and the absolute magnitude of a light source?

The actual signal time is only 0.01 sec. (signal>noise, after filters)

Can you help me with where this is in the paper - I can't find it.

Signal stretched out using Fourier and statistics 0.10 sec.

Can't find any reference to the signal being stretched out using Fourier and statistics - can you help me find that in the paper?

Frequency of actual signal: 100 Hz
Frequency of noise: 200 Hz.

Ditto - I can't find this either - can you help me to see where it is in the paper?

We can see again that the noise frequency is very similar to the signal.

Again, what are you looking at here?

In fact, in the end of the "wave", the signal continues, but is considered part of the noise.

Ditto, again.

It would be interesting to do good frequency analysis of the LIGO-system.
I suspect it can produce "chirp" signals from simple shocks or pulses.
Hint: the Fourier filter does not remove systematic noise.

Again, I'm not seeing this. What's a Fourier filter and how do you define systematic noise?

[Zyxzevn]

can you help me find that in the paper?

It is hidden in the actual data,
I was just reading the graph aloud.
Signal= lines following prediction.
Noise= lines not following prediction.

The paper does not do this analysis. They use statistics instead,
with a general frequency analysis.
If you know a bit of stochastic signal processing, you need white
noise (independent random noise) to do frequency analysis,
without having consequences.
But if you look at the raw data of the first GW claim, you can see that there
is a clear AM-signal in the noise. (AM=Amplitude Modulated)
This #%?$ up the whole frequency analysis, sadly.
You need to do a non-linear transformation to remove that AM signal first,
which is hard because it is very noisy.

The frequency analysis in the software is done with a common Fourier transformation.
This is on the LIGO website.
So that is why I call it a Fourier-filter.
The fourier-transformation converts the signal to frequency domain.
The filter removes the frequencies that are not interesting.
The reverse fourier-transformation converts the signal back to the time-domain.
The problem with this filter is that it adds new signals to the time-domain that
were not there before. For example a pulse, becomes a sinus wave after the filter.
This problem happens especially when there are non-linear noise elements,
like an AM-signal.

As I explained above somewhere, the "Chirp" and the AM-signal have similar
frequency components. This means that by converting everything to the frequency
domain, you may simply convert the AM-signal into a "Chirp".
And even the filtered signal, contains a lot of noise.
So the very chance is high that this is all a gross misidentification.

This AM-signal causes a real problem in their stochastic analysis.

And you are right that this NOT in the paper.
Because it should be.

[Zyxzevn]

Again, the real theoretical shape is not visible in their graph.
According to the real GW theory, the waves must gradually grow in frequency and amplitude together.

This animation shows what I mean...
http://imgur.com/gallery/TktgE0S

The GW theory signal is very long, while the signal of the LIGO is very short.
It is very very short if we use only the significant part of the signal
(the signal that is slightly bigger than the noise).

[Michael Mozina]

LIGO "found" a new gravitational wave.. (on January 4, 2017)

GW170104: Observation of a 50-Solar-Mass Binary Black Hole Coalescence at Redshift 0.2

https://journals.aps.org/prl/abstract/1 ... 118.221101

Redshift 0.2 is quite near, I think, and should be visible.

I have read the paper but I need some help with some things. First of all how far is z=0.2, say, in light-years? At whatever that distance is, what is the difference between the apparent and the absolute magnitude of a light source?

http://www.nature.com/news/ligo-spots-g ... Y-20170601

I think you'll find some of your answers in that article.

But the latest merger is the most distant detected so far. The gravitational waves it produced took somewhere between 1.6 billion and 4.3 billion years to reach Earth — most probably around 2.8 billion years. The results are published in Physical Review Letters.

It is supposedly more than twice as far away as the original signal, but they have quite a large fudge factor.

Again, what are you looking at here?

If you're looking at their methodology with respect to blip transients, the link below will explain it in detail. Essentially they claim to 'differentiate' between blip transients and black holes by *assuming* that blip transients cannot and do not ever increase in frequency very much, and that they cannot be observed by both detectors at once. The first assumption is the real "trick' in terms of the statistical manipulation. It allows them to stick blip transients in a different "bin' for purposes of mathematical manipulation.

https://dcc.ligo.org/public/0122/P15002 ... _burst.pdf

See section 2. The magic starts with their *assumption* that only black hole mergers can produce an increasing frequency pattern. The irony is that they chalk it up to what the author describes as an "ad hoc" variable that they simply added to their merger models. That's where all the statistical manipulation happens.

Again, I'm not seeing this. What's a Fourier filter and how do you define systematic noise?

Keep in mind that there are two types of "noise", environmental/extraneous noise which includes the full range of background interference from all environmental factors. The other "noise" that LIGO is using to manipulate the sigma numbers is related to *filtered* days and times, or "cherry picked" days and times that exclude all vetoed timelines too. It's essentially stripped of as much environmental noise as possible, but it's the "most clean" data they could find because it makes their sigma number higher.

Simply by virtue of the fact that they stripped out all the normal background interference precludes them from linking the *cause* of the signal to anything using the "filtered/statistical* set of data. All that mathematical calculation could possibly hope to demonstrate is that the signal is probably not a random fluke related to noise from *cherry picked days and times*. That's about all you could honestly say about that sigma calculation. A seven sigma number wouldn't tell us anything in terms of isolating the *cause* of the signal, even if it might demonstrate that the signal is real and it was caused by *something*.

The sigma manipulation begins with the way they filter the raw data. They're cherry picking the quietest days and only non-vetoed timelines to get the "cleanest" data so they can achieve the highest sigma number possible. It's all a ruse because the moment they strip out the normal background, their comparison to anything is irrelevant.

The sigma manipulate continues in the paper I cited earlier when they then *declare* that blip transients cannot be observed by both detectors simultaneously and they can't increase in frequency.

LIGO's approach to claiming a "discovery" of gravitational waves based on their cherry picked sigma number reminds me of that saying: Figures don't lie, but liars can sure figure.".

The whole 5+ sigma is a scam. It's a trumped up number just to get to the five sigma level so they can claim discovery. They had to cherry pick only 16 days from a total of 48 days just so they could manipulate the sigma to a high enough number.

FYI, my criticisms of the LIGO paper would *predict* that all similar 'discoveries' of "transient" gravitational waves by LIGO will be devoid of any support by any ground based or spaced based telescope or neutrino hardware. Today's announcement confirms my prediction. LIGO is already 0 for 3 at bat whereas I'm 3 for 3. Wanna bet that I go 4 for 4? What are the odds considering all the various hardware that we have in space and on the ground? Anyone want to trump up a sigma on the odds?

I shouldn't have to wait very long to find out if my predictions are correct. From the article:

As they presented the latest results at a press conference, LIGO researchers said that they are working on the analysis of at least six other ‘candidate events’, but declined to provide details.

[Cargo]

So Gravity is restricted by C now?

Where do they find the Time to come up with such complete nonsense.

[Higgsy]

can you help me find that in the paper?

It is hidden in the actual data,
I was just reading the graph aloud.
Signal= lines following prediction.
Noise= lines not following prediction.

So you eye-balled it. Right.

The paper does not do this analysis.

What analysis - eyeballing it?

They use statistics instead,
with a general frequency analysis.
If you know a bit of stochastic signal processing, you need white
noise (independent random noise) to do frequency analysis,
without having consequences.

What are you talking about? You need white noise to do what?

But if you look at the raw data of the first GW claim, you can see that there
is a clear AM-signal in the noise. (AM=Amplitude Modulated)
This #%?$ up the whole frequency analysis, sadly.
You need to do a non-linear transformation to remove that AM signal first,
which is hard because it is very noisy.

If you have an AM signal then the carrier will be at a particular frequency and easy to band filter out. But you're referring here to the original detection not this one?

The frequency analysis in the software is done with a common Fourier transformation.
This is on the LIGO website.
So that is why I call it a Fourier-filter.
The fourier-transformation converts the signal to frequency domain.
The filter removes the frequencies that are not interesting.
The reverse fourier-transformation converts the signal back to the time-domain.
The problem with this filter is that it adds new signals to the time-domain that
were not there before. For example a pulse, becomes a sinus wave after the filter.

And yet people use band pass filtering all the time in applications from physics to image processing to hi-fi

This problem happens especially when there are non-linear noise elements,
like an AM-signal.

In what way exactly is an AM signal "non-linear"?

As I explained above somewhere, the "Chirp" and the AM-signal have similar
frequency components.

What AM signal?

This means that by converting everything to the frequency
domain, you may simply convert the AM-signal into a "Chirp".

If it genuinely is an AM signal it won't chirp.

And even the filtered signal, contains a lot of noise.
So the very chance is high that this is all a gross misidentification.

And all that from eye-balling the data. Astonishing!

[Higgsy]

Again, the real theoretical shape is not visible in their graph.
According to the real GW theory, the waves must gradually grow in frequency and amplitude together.

Seems to grow in frequency and amplitude to me. What are looking at?

The GW theory signal is very long, while the signal of the LIGO is very short.
It is very very short if we use only the significant part of the signal
(the signal that is slightly bigger than the noise).

And yet the signal to noise ratio is 13. So the signal is a lot bigger than the noise, isn't it? Precisely how long should an inspiral and ring down be compared to what is measured?

[Higgsy]

LIGO "found" a new gravitational wave.. (on January 4, 2017)

GW170104: Observation of a 50-Solar-Mass Binary Black Hole Coalescence at Redshift 0.2

https://journals.aps.org/prl/abstract/1 ... 118.221101

Redshift 0.2 is quite near, I think, and should be visible.

I have read the paper but I need some help with some things. First of all how far is z=0.2, say, in light-years? At whatever that distance is, what is the difference between the apparent and the absolute magnitude of a light source?

http://www.nature.com/news/ligo-spots-g ... Y-20170601

I think you'll find some of your answers in that article.

But the latest merger is the most distant detected so far. The gravitational waves it produced took somewhere between 1.6 billion and 4.3 billion years to reach Earth — most probably around 2.8 billion years. The results are published in Physical Review Letters.

It is supposedly more than twice as far away as the original signal, but they have quite a large fudge factor.

And what would the difference in absolute and relative magnitude be for an event at 3 billion light years distance?

Simply by virtue of the fact that they stripped out all the normal background interference precludes them from linking the *cause* of the signal to anything using the "filtered/statistical* set of data. All that mathematical calculation could possibly hope to demonstrate is that the signal is probably not a random fluke related to noise from *cherry picked days and times*. That's about all you could honestly say about that sigma calculation. A seven sigma number wouldn't tell us anything in terms of isolating the *cause* of the signal, even if it might demonstrate that the signal is real and it was caused by *something*.

As I understand it, the sigma estimate is purely about the confidence that what they are observing is signal and not noise, and it says nothing and cannot say anything about the cause of the signal. Nor is it claimed that it does.

[Michael Mozina]

And what would the difference in absolute and relative magnitude be for an event at 3 billion light years distance?

Google is your friend:

http://www.1728.org/magntude.htm

As I understand it, the sigma estimate is purely about the confidence that what they are observing is signal and not noise, and it says nothing and cannot say anything about the cause of the signal. Nor is it claimed that it does.

Then what exactly makes this a "discovery" and what does *that* sigma calculation have to do with putting this signal it into the "discovery" category if the *cause* of the signal can't be determined from that sigma? I'm definitely *not* impressed with their fuzzy sigma. I see *zero* empirical evidence to support their claim as to cause and I see no relevant math to support their claim of "discovery". All they seem to have "discovered' is a clever way to fudge the sigma.

[Higgsy]

And what would the difference in absolute and relative magnitude be for an event at 3 billion light years distance?

Google is your friend:

http://www.1728.org/magntude.htm

...and therefore the difference bewteen absolute and apparent magnitude at 3 billion light years is 40 which is a huge difference.

As I understand it, the sigma estimate is purely about the confidence that what they are observing is signal and not noise, and it says nothing and cannot say anything about the cause of the signal. Nor is it claimed that it does.

Then what exactly makes this a "discovery" and what does *that* sigma calculation have to do with putting this signal it into the "discovery" category if the *cause* of the signal can't be determined from that sigma? I'm definitely *not* impressed with their fuzzy sigma. I see *zero* empirical evidence to support their claim as to cause and I see no relevant math to support their claim of "discovery". All they seem to have "discovered' is a clever way to fudge the sigma.

Sigma always relates to the confidence that what is being observed is signal rather than noise. Not just here in LIGO but in general. It never says anyrthing about the cause of a signal. In this case as I understand it, all the systematic and environmental causes have been eliminated, and the signal closely matches that expected for a black hole merger, including both the inspiral and ring-down. Hence the conclusion that the cause is a black hole merger.

[comingfrom]

Thank you, Higgsey.

In this case as I understand it, all the systematic and environmental causes have been eliminated, and the signal closely matches that expected for a black hole merger, including both the inspiral and ring-down. Hence the conclusion that the cause is a black hole merger.

Most of us have read the paper. Since black holes are as yet an unproven theory, and in fact are proven (in many ways) cannot possibly exist, that eliminates black holes too.

For those of us that cannot believe in black holes, we see that LIGO has constructed it's conclusion on presumption upon presumption. Presumptions which not all of us hold to be true.

Belief in LIGO's conclusion first requires a dogmatic belief in the existence black holes, something which has not been been verified, and which arose from the need to fill in gaps in theory, and was created by dividing by zero.

If their paper is approached with disbelief, the absurdity of it becomes even more apparent. It is claimed that the LIGO instruments are sensitive to movements less than the diameter of a proton. So how did they eliminate the movement of the molecules within the mirrors, which reflect the lasers? Of course, this is not mentioned. Maybe it is also presumed that molecules have no freedom of movement within solid substances, and that the laser remains focused directly on the center of a proton in a molecule in the mirror, until the mirror moves.

Furthermore. Clearly, they can only have instruments that measure the known systematic and environmental causes, in order to eliminate them. Saying merging black holes is the only other possible cause is presuming there are no other unknown causes. But since unknown causes are unknown, it cannot be said with any confidence that all possible causes have been eliminated. Pure logic.

Maybe galactic central burped. Did they eliminate that possibility? Do we even measure expulsions from galactic central yet? Maybe it was [what is called] a magnetic reconnection, high above LIGO, where no satellite was to report it. What I mean is, EU proponents can easily think of possible (or at least plausible) known causes which didn't get eliminated. How many unknown causes there might be is not and cannot be known.

One thing I do know, it wasn't black holes.
It might even have been a gravitational wave (though I also think that is highly unlikely), but it still definitely wasn't caused by black holes.
~Paul

[Zyxzevn]

Astonishing!

It is indeed astonishing that by just looking at some data, one can already
see the grave errors that these scientists have made in their analysis.

It is astonishing how far these scientists have come from understanding
the basics of signal versus noise.

And it is astonishing how they did not look at the resonating noise signals in
their system, that #%?$ up their signal with an Amplitude modulation.

It is astonishing that these scientists use a basic frequency analysis, without
knowing that it makes no sense in stochastic signal theory, with these kinds of signals.

It is astonishing that many other scientists just swallow this analysis without much
criticism.

It is astonishing how this all has a high sigma value.

It is as much science as the face of Mars.