quentron.com

[sarkeizen]

Just the same, I would be happy to help him work out solid test protocols if he wants my help.

My advice would be to make them as strict as possible.  The mathematical reason behind statements like: "Extraordinary claims, require extraordinary evidence" is the simple fact that given a test with a false positive rate of 1 in 1 000 000 and two hypotheses H1 (very likely to be true P(H1) = 0.8 ) and H2 (very likely to be untrue P(H2)= 0.0000000000000000000000000000000000000000000000000000000008).

H1 has a 1 in 2 000 000 chance of getting a false positive.
H2 has approximately a 1 in 1 000 000 chance of getting a false positive.

So evidence for H2 is worth only *half* the evidence for H1.

[MarkE]

My advice would be to make them as strict as possible.  The mathematical reason behind statements like: "Extraordinary claims, require extraordinary evidence" is the simple fact that given a test with a false positive rate of 1 in 1 000 000 and two hypotheses H1 (very likely to be true P(H1) = 0.8 ) and H2 (very likely to be untrue P(H2)= 0.0000000000000000000000000000000000000000000000000000000008).

H1 has a 1 in 2 000 000 chance of getting a false positive.
H2 has approximately a 1 in 1 000 000 chance of getting a false positive.

So evidence for H2 is worth only *half* the evidence for H1.

A lot of people have trouble with the (1 - N) that frequents statistics.

[Philip Hardcastle]

I think the only sensible reply to people trying to ascribe probabilities that have no actual knowledge is that of the PM Benjamin Disraeli;


"Lies, damned lies, and statistics"

"it is a phrase describing the persuasive power of numbers, particularly the use of statistics to bolster weak arguments. It is also sometimes colloquially used to doubt statistics used to prove an opponent's point."


I cannot for strict confidentiality reasons divulge anything on the topics of what is to be released in August, so I will use the British Royal position of making no comment, but with a slightly cheeky smile.

As I stated before I am happy to debate M Demons on the Karpen thread, but not here.

Thanks.

[sarkeizen]

I think the only sensible reply to people trying to ascribe probabilities that have no actual knowledge

Since we're back to not actually speaking my name...which is kind of cool it makes me feel like some kind of diety or Lovecraftian mythos.  I'll assume you're talking about me.  Every prediction I've made about your producing quenco has been correct.  Right?  Not one wrong prediction from me.  Period.  100% correct. Right? Every prediction on the same issue, by you has been wrong.  Right? 100% There is not one working quenco. 

So remind me again, which one of us has no actual knowledge?

is that of the PM Benjamin Disraeli;

Because it's always good to get your advice on mathematics from politicians and amateurs instead of actual mathematicians.

I will use the British Royal position of making no comment, but with a slightly cheeky smile.

...to cover up your inevitable failure.  Were I giving odds that you were a charlatan, your ability to adopt the same ridiculous posturing every summer for the past what three? four? years.  Would shift the odds in favor of that conclusion.

As I stated (stupidly) before I am happy to debate M Demons on the Karpen thread (which has nothing to do with Maxwell's Demon), but not here (on a thread where I claim to have constructed a Maxwell's Demon)

For once Philip, you and I are on the same page.

Thanks.

[MarkE]

Mr. Hardcastle statistics can be very helpful to those who want to find the truth.  The fact that some have abused figures by misapplying them doesn't change the utility of statistics.  It merely tells us that we must carefully evaluate the logic used and actual evidence provided in support of any argument.  If you have intellectual property concerns that prevent open discussion of test protocols, then I hope you will do whatever you can to vet your proposed test protocols with qualified persons in private.