I have found throughout S2 that I have confusion with when and when not to use the unbiased estimator for the variance.
I understand if you have a sample from a population of UNKNOWN variance, you must use it then. However, there seems to be discontinuity in some questions. Sometimes I have noticed that with this question: (taken from June 2012 S2 q2 i):
For the continuous random variable V, it is known that E(V) = 72.0. The mean of a random sample of 40 observations of V is denoted V bar. Given that P(V bar < 71.2) = 0.35, estimate the value of Var(V).
Now I managed to get Var(V) before applying the unbiased estimator. I then applied the unbiased estimating method (multiplying by the sample size/ sample size -1) and that was not the answer.
However, when you are given data for the sample of a population (given the summation information) you have to use the unbiased estimator.
I am not entirely sure why, but can someone please explain this to me.
[OCR not MEI] Statistics 2: Deciding on when to used an unbiased estimator
|TSR's new app is coming! Sign up here to try it first >>||17-10-2016|