Probability Through the Lens of Coin Flipping

[marketsurfer]

http://labs.elmfunds.com/pastreturns?fwd=cd&data={"emails":{},"phones":{},"first_name":"Dave","timezone":"Europe/London","title":"Specialized Financial Writing","email":"marketsurfer@gmail.com","name":"dna","company":"DNA","last_name":"Goodboy"}

When are Past Returns Indicative of Future Returns?
A Brief Exploration Through the Lens of Coin Flipping

You are presented with two coins, one is fair, and the other has a 60% chance of coming up heads. Unfortunately, you don't know which is which.2

What number of flips would you want to see performed in parallel on the two coins to give yourself a 95% chance of identifying the biased one? Please just give it a quick guess, without working it out on a pad of paper. You need to hit 'answer' in order to read the rest of the note.

 

[drm7]

I'm guessing 60. Only basing this on my experience at a credit card company, which needed a minimum of 30 credit approvals per risk bucket to adequately calibrate the model. That was the bare minimum though.

 

[southall]

Well i guessed 200.
Traders dont need to wait years, we have daily results to use.
So can get high confidence of statistically significant results much easier and quicker.

 

[vanzandt]

There's no correct answer.

 

[southall]

There's no correct answer.

It asked for an estimate for 95% confidence, not 100% confidence.

 

[Xela]

There's no correct answer.

That's certainly not right: it's a simple mathematical function, so it has an answer.

The link isn't working at the moment, but my guess is about 150, because 60% is pretty close to 50%

 

[Option_Attack]

Hmmm...if you could factor in the probability of Surfer's YM trades ending up in the green...sky's the limit!

 

[Visaria]

500 should give me 95% confidence.

 

[JackRab]

>90

 

[JackRab]

You do know that this post doesn't really help your money management business @marketsurfer? Since it basically says, don't give your money to active managers