Hypothesis Test question

[emonrad87]

I am doing a hypothesis test for the probability of being dealt a certain hand compared to my actual results (this is for a statistics class). So, I use the equation:

z = (P - Po) / sqrt((Po(1-Po))/n)

My question is, what's my n? Is it the total number of hands seen, or is it the number of times I've seen that particular hand?

[gaming_mouse]

n is the total # of hands seen. Po is the true prob of being dealt the hand in question.

This test doesn't make much sense though. What are you trying to prove? That there is evidence that there is some bias for or against getting this hand? I can't think of any other use for a test here.

gm

[dkernler]

It just sounds like a question concerning the frequency of the particular hand. It's a Central Limit Theorem question, really. Over a particular number of hands, we'd expect Po*n of them, but it's rarely exactly that. So if we happened to have a significantly larger (or smaller) number, it might be interesting to see how likely that number is.

I will say this - I'm not sure why you're studying this in a course. It's certainly not something I'd assign in mine. What level of Statistics are you studying?

[gaming_mouse]

[ QUOTE ]
Over a particular number of hands, we'd expect Po*n of them, but it's rarely exactly that. So if we happened to have a significantly larger (or smaller) number, it might be interesting to see how likely that number is.

[/ QUOTE ]

Yes, I got that. My point is, unless you think online poker is rigged, there is no real point to the test. But whatever... it's a fine exercise.

gm

[dkernler]

[ QUOTE ]
Yes, I got that. My point is, unless you think online poker is rigged, there is no real point to the test. But whatever... it's a fine exercise.

[/ QUOTE ]
You're exactly right. It's really pointless - only an exercise. Didn't mean to offend.