Thought experiment for you "do I have a winning model" types

[mathdotcom]

Suppose I have 100 quarters in a jar and I wonder if they are unfair in the sense that they have an inherent tendency to land on heads or tails significantly more than 50% of the time. I know the coins are all identical. I hire 100 people to go home with one coin each, and ask them to flip their coin 200 times and record the results. I tell them to come back to me only if they think they have an unfair coin. Each of these 100 people does not know I have hired 99 others to do the same.

Suppose one of the 100 people I hired, let's call him Mr.Ed, comes back to me and tells me his coin landed 130 heads and 70 tails.

1. What is the probability in the head of Mr.Ed that he has an 'unfair' coin?
2. What is the probability in my head that Mr.Ed's coin is unfair?

I am not asking for answers. I want you to think about the answers in terms of relative values not absolute values.

This public service announcement brought to you by MDC Industries.

[hutennis]

Is there a reason why relative and absolute (whatever that means) values or his or yours probability should differ in this case?
Where would any kind of difference come from if you trust his report.
It is one experiment independent of anything that produced independent results.
What psychological pitfalls are you referring to?

[mathdotcom]

Is there a reason why relative and absolute (whatever that means) values or his or yours probability should differ in this case?
Where would any kind of difference come from if you trust his report.
It is one experiment independent of anything that produced independent results.
What psychological pitfalls are you referring to?

Uhh how about the two agents' different amounts of information?

[hutennis]

Different amount of what information?

[hutennis]

LOL.

Let me rephrase it.

How does any information you may have that Mr.Ed does not changes physical properties of that coin making it to stop obeying Law of gravity?
How does any information you may have that Mr.Ed does not allows you to interpret physical results
any different than bell curve does?

[mathdotcom]

LOL.

Let me rephrase it.

How does any information you may have that Mr.Ed does not changes physical properties of that coin making it to stop obeying Law of gravity?
How does any information you may have that Mr.Ed does not allows you to interpret physical results
any different than bell curve does?

Apparently you've never heard of an unfair coin that can be significantly biased towards heads while still obeying the laws of gravity

[hutennis]

Apparently you've never heard of an unfair coin that can be significantly biased towards heads while still obeying the laws of gravity

Thats my point.

Fair coin will do what fair coin does.
Unfair coin will do what unfair coin does
CLT will estimate corresponding probabilities.

What kind of different amount of agent's information are you talking about that can in any way impact
interpretation of results?
Are you high?
Or you just trying to play Kahneman and Tversky overhere?

[mathdotcom]

Just stop posting

Please

[hutennis]

Just stop posting

Please

Why?
What would the reason now?


Why dont you just answer the question?

What kind of different amount of agent's information are you talking about that can in any way impact
statistical interpretation of the results?

[mathdotcom]

For one your english (or your brain) is so dysfunctional I don't even know what you're trying to communicate

[hutennis]

For one your english (or your brain) is so dysfunctional I don't even know what you're trying to communicate

Or you just got caught spreading nonsense and are trying to get out of embarrassing situation using your favorite method -
good old ad hominem, aka personal attack.

Here is your argument that supposed to somehow open the door to different interpretations of probabilities in your scenario:

Uhh how about the two agents' different amounts of information?

Here is my question to your argument:

What kind of different amount of agent's information are you talking about that can in any way impact
statistical interpretation of the results?

Who else around here thinks that my question is irrelevant or incoherent?

Answer the question please.

[mathdotcom]

When I say probability I mean p-value from an appropriate hypothesis test

The point of the thread was for you to answer that question for yourself

I am sorry you can't

And yes you are a retard and yes that is my favorite, ad hominem which you have so kindly translated for me about 30 times

[hutennis]

LOL

p value IS a probability. It measures the chance that fair coin would give a result at least as extreme as observed.

130 heads out of 200 flips gives us one-sided p value of 0.00001.

Now, how different amount of any kind of information agents (yuou and Mr.Ed) may have can make this probability to be different in MR.Ed's head than in yours?

[mathdotcom]

LOL

p value IS a probability. It measures the chance that fair coin would give a result at least as extreme as observed.

130 heads out of 200 flips gives us one-sided p value of 0.00001.

... that's what I said. I just tried to be more specific since your english is horseshit. I was trying to make sure you understood I was talking about a p-value instead of a 'true probability'.

Now, how different amount of any kind of information agents (yuou and Mr.Ed) may have can make this probability to be different in MR.Ed's head than in yours?

That can make kind of question for hutennis think maybe understand about of

[hutennis]

BTW, speaking about retards.

How would you call someone who'd come up with "thought experiment"as retarded as yours?

[mathdotcom]

BTW, speaking about retards.

How would you call someone who'd come up with "thought experiment"as retarded as yours?

There you go again with your ad hominem attacks (ie personal attack, like to insult a person)

I'm sorry you can't figure it out

[hutennis]

That can make kind of question for hutennis think maybe understand about of

Please, explain this then

Uhh how about the two agents' different amounts of information?

[mathdotcom]

I know that 99 others did not get back to me. MrEd does not know this.

[thom321]

This thread is pure (unintentional?) comedy.

[mathdotcom]

[hutennis]

I know that 99 others did not get back to me. MrEd does not know this.

OK, so how is that supposed to make any difference for answering your 2 questions?

1. What is the probability in the head of Mr.Ed that he has an 'unfair' coin?
2. What is the probability in my head that Mr.Ed's coin is unfair?

No matter what you know that MrEd does not, unless one of you is retarded
the answer has to be the same - 0.99999.

you can think about the answers in terms of relative values or absolute values
untill you are blue in your retarded face, the answer is still the same

[mathdotcom]

hutennis

Suppose I know the coins are fair

How many do you expect will come back to me and tell me they found p-values of 0.05 or lower?

[mr.ed]

Mr. Ed, if he is hip to probability, will be virtually certain that there is something wrong with the coin.

You, on the other hand, will know that nothing is wrong with the coin(s).

[hutennis]

hutennis

Suppose I know the coins are fair

How many do you expect will come back to me and tell me they found p-values of 0.05 or lower?

If you "know" that coin Mr.Ed was flipping is fair, I highly recommend you to check your testing methods.
They must be kinda retarded.

[mathdotcom]

MrEd !!!!



hutennis....

[hutennis]

I'm still waiting for at least half-retarded explanation on what all this relative value, absolute value and different probability in different heads nonsense supposed to mean.

If your smiling faces collage indicates that this

Mr. Ed, if he is hip to probability, will be virtually certain that there is something wrong with the coin.

You, on the other hand, will know that nothing is wrong with the coin(s).

is what you really mean, you should go to the doctor as soon as possible.

[thom321]

Someone should definitely see a doctor....

[mathdotcom]

I agree with the general logic behind what he said.

Here is an example of speaking in relative terms:

Jose Canseco weighs more than Kate Upton

Speaking absolutely,
Jose Canseco weighs 250lbs and Kate Upton weighs 115lbs.

MrEd is very confident he has an unfair coin, but I do not share his confidence since someone was due to get skewed results. The fact that only MrEd got back to me tells me the other 99 got basically "even" results. I therefore have more information and that is why I have a different inference about the coins.

This is why HTT posters are generally skeptical of new posters claiming they've found a system.

I will book doctor's appt first thing tomorrow morning.

[hutennis]

I'd say so.

If someone thinks that knowing about bunch of coins being flipped somewhere and thinking in relative terms about the coin that delivered 130 heads in 200 flips can somehow make this coin fair, a prompt doctor appointment is in order.

[mathdotcom]

I'd say so.

If someone thinks that knowing about bunch of coins being flipped somewhere and thinking in relative terms about the coin that delivered 130 heads in 200 flips can somehow make this coin fair, a prompt doctor appointment is in order.

My point was never to discuss absolute p-values, which you insisted on doing. You can answer the question without doing that.

But since you insist on doing that, what if MrEd comes back to me with 114 heads out of 200?

Should I infer the coins are fair or unfair?

[thom321]

I absolutely glad I continued to read this thread since it got relatively funnier the longer it went on.

[hutennis]

This is why HTT posters are generally skeptical of new posters claiming they've found a system.

HTT poster should be skeptical, or, better yet, disregard completely any claim about any system/model
found by new or old poster b/c probability is 0.99999 that this system/model is nothing more but
past data collection put together using merciless over fitting.

[hutennis]

But since you insist on doing that, what if MrEd comes back to me with 114 heads out of 200?

Should I infer the coins are fair or unfair?

In this case, the probability that his coin is fair is 0.028, regardless of how many millions of other coins are being flipped for you around the world and how many millions of other results you expect to receive.
Whether you think about results obtained for you by MrED in relative or absolute terms does not make any difference either. Probability of his coin to be fair is still the same - 0.028.

[MonkeyF0cker]

HTT poster should be skeptical, or, better yet, disregard completely any claim about any system/model
found by new or old poster b/c probability is 0.99999 that this system/model is nothing more but
past data collection put together using merciless over fitting.

Are you too dumb to realize that you're agreeing with Mathy?

[hutennis]

Are you too dumb to realize that you're agreeing with Mathy?

Are you too dumb to realize the subject is changed?