Hi, I've come across a question which states something like: Sample size of 20, sample mean of 2000 and sample standard deviation of 500. The distribution is log normal (mu, sigma^2) I have to obtain the method of moments of mu and sigma^2. Its driving me mad. I don't see how I need to use the sample size of 20 at all? Surely its just: Equate the sample mean with the population mean to get the first equation. Then equate the sample s.d. squared with the population variance to get the second equation which is: 2000^2 (e^(sigma^2)-1) = 500^2 => sigma ^2 = 0.06 => mu = 7.57 Is this correct or should the sample 20 value be used in the sample s.d. somewhere? Thanks.
Yes this is exactly the answer i attained for this question. I don't think the question setter put the n=20 in there maybe for confussion purposes, because x bar, and s are given in the question. The only time this info would be useful would be if Sum xi, and Sum(x-xi)^2 were given in the question instead, but this is not the case, and i think either this was a clever way to confuse people, or the person who wrote the question didn't realise this info is not required but put it there anyway.
