Statistics question: a binomial model can be approximated with what other model?

Normal distribution?

Well, we’re really not supposed to give away homework answers on Fluther, but I do know this one. I’m trying to think how to lead you to it without just telling you.

When you have very large samples sizes, there is a type of distribution that begins to behave similarly to a binomial distribution. There are some formulas: for large n, μ ≈ np and σ ≈ √(np(1-p)).

Got any ideas now?

I want to say normal distribution, but I’m still not 100% sure.

Normal model*

Right, so how can you be more confident in that answer? Look at my formulas. These are the formulas for approximating a binomial distribution: μ ≈ np and σ ≈ √(np(1-p)). n and p are variables you see in binomial distributions, right? So the approximation must use μ and σ. Which distribution do you know that uses those variables? You’re right – in the normal distribution mu is the mean and sigma is the standard deviation.

AHA! Thank you so much, makes perfect sense now! You should be my Stats teacher! :-P