Let's assume I have a relatively large sample (500) from an unknown distribution. I can calculate sample mean and variance, can I calculate confidence intervals for them? I've googled and found that because of the central limit theorem it should somehow work out for the mean (what distribution would I use in the formula tho? Normal? T?)
What about variance?
Also stats thread or whatever.
Relax, if I understood everything I've read I wouldn't post here, would I? I'm shit at stats (and don't really understand confidence intervals), so I was hoping someone could explain this to me. Clearly, you're not the person for the job?
>In probability theory, the central limit theorem (CLT) states that, given certain conditions, the arithmetic mean of a sufficiently large number of iterates of independent random variables, each with a well-defined expected value and well-defined variance, will be approximately normally distributed, regardless of the underlying distribution
Okay, so how does that answer my question? From this I can learn that the average value is approximately normal if I have a large enough sample. Is that sufficient enough reason to use the z table? What about variance/standard deviation?
Your problem, as you already realize, is that you know nothing about the distribution of the population from which your sample has been selected. Google bootstrapping. You're welcome...
Why does everyone here always kick up dust when they can't answer simple questions succinctly? Even if you knew what you're talking about, you're not going to impress some beginner who can't even into CIs with nuanced arguments about assumptions...
Use a t-test because you don't know the population standard deviation.
Fucking thank you. You mean Student's t distribution, not t-test (I don't want to test hypotheses), I presume?
Am I to assume these formulas I've found are incorrect?