Hi everyone! I need your help, if you have a moment. It's rather a standard problem in statistics, but I'm not much familiarized.
Assume that we draw samples from a normal distribution $N(\mu,\Sigma)$, whose parameters are not known. How many samples should one generate in order to obtain a good estimation of the sample mean $\mu=\frac{1}{N}\sum_{i=1}^{n}x_i$? How is this related to the dimensionality of the samples?
Thank you very much!