111–120 of 123 posts
Has anyone investigated Dean Radin's work? It demonstrates similar statistical effects in a huge number of experiments. He has put forth quantum entanglement as the explanation.
http://en.wikipedia.org/wiki/Funnel_plot
PS non-physicists invoking "quantum stuff" are bullshit merchants, p < 0.001 :-)
Earlier quoted context omitted.
This is why we replicate studies. Unreplicated parapsychology results are a dime a dozen.
Who are "we"? Physicists are the only ones who still do as far as I can tell -- everyone else prefers to spend their research budget somewhere else.
The whole field of systematic reviews and meta-analyses has developed around the need to aggregate results from multiple studies of the same disease or treatment, because you can't just trust one isolated result -- it's probably wrong.
http://en.wikipedia.org/wiki/Meta-analyses
Statisticians working in EBM have developed techniques for detecting the 'file-drawer problem' of unpublished negative studies, and correcting for multiple tests (data-dredging). Other fields have a lot to learn...
Earlier quoted context omitted.
Having moved from physics to biology, I am amazed with the difference in what the consensus of 'significant' is. Some of the difference is due to necessity, but not all.
There are more reasons to doubt our results, so we lower our standards of evidence?
When some people find that their model doesn't quite fit, they make a more accurate model. Others make a less specific model. It's the difference between model parametrization and model selection.
So when we get a dubious result, we can either say "no result" or "possible result". The choice tends to depend on how the finding affects future research. Biology is more exploratory than confirmatory, so they go that way.
Earlier quoted context omitted.
Unfortunatley, publishing these kind of claims prematurely help the more gullible among us to fall for ridiculous claims from psychics and others who would take advantage of them. True. But on the other hand, publishing ridiculous claims and incorrect results is a necessary part of science. When we publish only results we know to be correct, because they agree with mainstream beliefs, we introduce a bias into the sci…
> In reality, if you publish 20 experiments with p=0.05 [1], 1 of them should be incorrect. In reality it doesn't turn out this way because the results that get written and published tend to be biased in favour of novelty and demonstrating a relationship rather than the absence of one. How many similar experiments could have been terminated, never submitted or not published because they failed to show anything notabl…
On a similar note, how many times did a person choose the left or right curtain? Was that 50/50?
I can imagine the subject choosing the left or right more often based on the environment. For instance, the subject is sitting in an asymmetric position usually (mouse on right side). Was this considered when taking the test? Was the subject left or right handed?
You can look at numbers all you want but there are always biases to take into account and figuring out why we have those biases might be more important than this study.
Earlier quoted context omitted.
So far the best line in this paper is: "Returning to Laplace’s Principle, we should obviously assign our prior belief in precognition a number very close to zero, perhaps slightly larger than the probability of, say, goldfish being able to talk"
You touch upon a very important point. For some reason we are very bad at Bayesian reasoning. Sometime it leads to unsubtantiable opinions http://news.ycombinator.com/item?id=1909576 but could be very damaging if faulty reasoning is used by a jury or a judge...."the person can be guilty or not guilty so the probabilities are 50:50". I would be a lot more relieved if people with the power over other peoples lives grok…
Bayes's rule doesn't help with the point that suggestive evidence is not convincing evidence. It just points out that prior beliefs are part of the equation, but will hopefully pale in comparison to actual data. In fact, I was taught to set practically useless hyperparameters to ensure that they do. No one does that outside of an experiment.
Let's say I believe (I don't) the height of pygmies is normally distributed, where the mean is also normally distributed with mean 130cm and standard deviation 10cm, and the standard deviation is inverse gamma distributed with shape 7cm and scale 1cm. Assuming the height is actually normally distributed with mean 160cm and sd 15cm (it isn't), how many pygmies must I measure to admit that P(height>160cm)>20%? I'm not sure I can even do the math.
Here P=50% for the unknowable accurate model and P=0.13% for the prior model. How does the situation change when my prior is "sufficiently close to 0"?
Earlier quoted context omitted.
Who are "we"? Physicists are the only ones who still do as far as I can tell -- everyone else prefers to spend their research budget somewhere else.
Clinical researchers too. Because lives are at stake. The whole field of systematic reviews and meta-analyses has developed around the need to aggregate results from multiple studies of the same disease or treatment, because you can't just trust one isolated result -- it's probably wrong. http://en.wikipedia.org/wiki/Meta-analyses Statisticians working in EBM have developed techniques for detecting the 'file-drawer p…
Regardless of the true reason, these are never carried out before a new drug or treatment is approved (because there is usually one or two studies supporting said treatment, both positive).
And if you have pointers to techniques developed for/by EBM practitioners, I would be grateful. Being a bayesian guy myself and having spent some time reading Lancet, NEMJ and BMJ papers, I'm so far unimpressed, to say the least.
Earlier quoted context omitted.
Oh big deal, who cares if some people have their faith in the paranormal strengthened. Something like one in two people or more believe in what scientists would consider paranormal already. Good scientists know there is much yet to be discovered.
The problem is that these people are allowed to vote and hold office.
Earlier quoted context omitted.
You touch upon a very important point. For some reason we are very bad at Bayesian reasoning. Sometime it leads to unsubtantiable opinions http://news.ycombinator.com/item?id=1909576 but could be very damaging if faulty reasoning is used by a jury or a judge...."the person can be guilty or not guilty so the probabilities are 50:50". I would be a lot more relieved if people with the power over other peoples lives grok…
Please explain the proper way to form priors, then. 50:50 is widely used, as is "0 for obviously wrong stuff". The author here suggested "something sufficiently close to 0", which to me is indistinguishable from the second one. Should an accused's guilt prior be based on the jury's own guilt, the number of crimes they've heard about recently, or the judge's conviction rate? Or maybe the accused's socio-economic class…
In the legal case example, maybe some clarity maybe had in considering what does the prior mean. A answer is: say you have to bet a million dollars on whether the person is guilty or not without knowing anything about the person how would you distribute your million dollars between the two events. Yes it is subjective and personal, but it is hardly ever going to be 50:50. One can push the $1,000,000 analogy further. One can fix a cost for a mistake: whats the cost of a wrong conviction and whats the cost for setting a guilty man free. Then the final decision can be based on reducing the financial risk based on the likelihoods.
One may bring the socio-economic status in forming the priors but one may not consider any information source that considers the accused.
Earlier quoted context omitted.
You touch upon a very important point. For some reason we are very bad at Bayesian reasoning. Sometime it leads to unsubtantiable opinions http://news.ycombinator.com/item?id=1909576 but could be very damaging if faulty reasoning is used by a jury or a judge...."the person can be guilty or not guilty so the probabilities are 50:50". I would be a lot more relieved if people with the power over other peoples lives grok…
Please explain the proper way to form priors, then. 50:50 is widely used, as is "0 for obviously wrong stuff". The author here suggested "something sufficiently close to 0", which to me is indistinguishable from the second one. Should an accused's guilt prior be based on the jury's own guilt, the number of crimes they've heard about recently, or the judge's conviction rate? Or maybe the accused's socio-economic class…
Lets take the example. There is a one to one correspondence with fictitious counts and priors. One way of encoding a 50:50 prior is to construct a possibly fictitious but representative past of (say) 2000 samples split into 1000 guilty and a 1000 not-guilty. After each prediction and assuming that the truth gets known one has to update the counts appropriately, so that the next time we use a different prior.
Our initial prior may be wrong but it will approach the correct one asymptotically. But how fast it approaches the true prior depends on how wrong our initial prior was.