Let's make a deal.

[TQM]

Just a little puzzle for you (an oldie to be sure):

The honest game show host shows you 3 doors. He tells you behind one of the doors is a Mercedes SLR. Behind each of the other doors is a goat.

He asks you to choose which door you want and he tells you after you've done so, he'll reveal a goat behind one of the remaining unchosen doors.

After he reveals the goat behind one of the other doors, he gives you the choice of switching your choice from the door you've chosen to the door that's remaining.

Do you switch? Or, do you keep with your original choice? What's the best thing to do?

 

[ed_v]

I'd go for the goat....I've already got a Mercedes

 

[papasmerf]

so I get to keep the goat and no one will guestion me??????

 

[mmouse]

Trick question! Some people prefer a goat than a Mercedes.

 

[TQM]

point of order

It wasn't just any mercedes - it was an SLR.

why that should matter,.....

 

[mmouse]

Because if you want the goat you shouldn't switch.
http://en.wikipedia.org/wiki/Monty_Hall_problem

 

[TQM]

Thanks for being an idiot.

I'm fully capable of looking up websites.

 

[red]

i know remember why I hated stats

 

[someone]

Just a little puzzle for you (an oldie to be sure):

The honest game show host shows you 3 doors. He tells you behind one of the doors is a Mercedes SLR. Behind each of the other doors is a goat.

He asks you to choose which door you want and he tells you after you've done so, he'll reveal a goat behind one of the remaining unchosen doors.

After he reveals the goat behind one of the other doors, he gives you the choice of switching your choice from the door you've chosen to the door that's remaining.

Do you switch? Or, do you keep with your original choice? What's the best thing to do?

This is a classic from both statistical theory and the economics of uncertainty.. If you’re a Bayesian updater you switch.

 

[red]

This is a classic from both statistical theory and the economics of uncertainty.. If you’re a Bayesian updater you switch.

so bayesian updaters are switchs? i didn't know they were in bdsm. i guess i really never paid attention in class

 

[1HandInMyPocket]

This is a classic from both statistical theory and the economics of uncertainty.. If you’re a Bayesian updater you switch.

I believe the correct statistical answer is to not switch. Say you choose door #1, then there is a 67% chance the goat being behind doors 2 and 3. Say he reveals that door #2 has a goat. Now door #3 has a 67% chance of having the goat, not 50%.

 

[thompo69]

I believe the correct statistical answer is to not switch. Say you choose door #1, then there is a 67% chance the goat being behind doors 2 and 3. Say he reveals that door #2 has a goat. Now door #3 has a 67% chance of having the goat, not 50%.

Click on the link in post 7. It explains the stats...

 

[yippie]

I believe this question (or a version thereof dealing with a ball and three cups) first appeared in a mathematics journal (there apparently are such things) in the 1980s and caused quite a stir as many of the world's top mathematicians got the answer wrong.

To a student of statistics, the answer that immediately comes to mind is "It doesn't matter". There are two doors left, the Mercedes is behind one of them and therefore you have a 50-50 chance of guessing right. However, probability theory is entirely based on randomness. The key is that when the game show host chose which door to open, he did not choose it at random. The door had to meet two conditions: It could not be the door the contestant had chosen and it could not have the Mercedes behind it. Because of this, when one door was opened it did not change the probability that the contestant's initial choice had been right - it was still one in three. What the contestant learned was that if his initial guess was wrong (probability two in three), the Mercedes had to be behind the one remaining door he had not chosen. Therefore he should change his mind.

Had the game show host opened a door at random, it so happened that the Mercedes was not behind it, and had THEN asked the contestant if he wanted to change his mind, then it wouldn't have mattered. Then the probability his intial guess was correct would have been one in two.

 

[1HandInMyPocket]

Click on the link in post 7. It explains the stats...

saw the link but didn't click it because I thought he was showing a link to a Monty Python skit. I guess I remember the solution wrong. I remember my prof trying to explain in vain why it is not a 50-50 probability.

 

[Dong Joe]

When you made your first pick you were 1 in 3 to pick the benz.

When he showed you where a goat is and you switch, you're 1 in 2.

Only by switching your choice can you increase the odds.

 

[frankcastle]

Yup this is a classic stats problem.... you switch and you have a better chance of winning. OFten called the Monte Hall Question.