I know this has been asked and answered, but many even have contradictory definitions as to which represents the mean of a probability distribution. I would like to cite: https://stats.stackexchange.com/a/137397/296060.
Regarding this, I would like to know: In the second case of un-fair die we have only taken a sample of 6 outcomes {1, 2, 3, 4, 5, 6}, therefore Sample Mean and E[X] are different. If we had taken a larger sample, the probability of each term would act out and, the Sample Mean and E[X] would again be the same, am I right?
Eg: A sample of 15 outcomes instead of 6 would have given {1, 1, 2, 2, 3, 3, 4, ,4, 4, 5, 5, 5, 6, 6, 6}, right ?
And then Sample Mean and E[X] would be same = 3.80.
Am I correct ?

  • $\begingroup$ Welcome to Cross Validated! What is $X$ in your mind? Traditionally, we would consider the sample mean to be an estimate of the expected value of a random variable $X$, where $X$, loosely speaking, is the population distribution. $\endgroup$ – Dave Sep 10 '20 at 3:00
  • $\begingroup$ Yeah. And so, can we say that a good sample is one where the E[X] is or almost equal to sample mean ? $\endgroup$ – Divyansh Singh Sep 10 '20 at 3:16
  • $\begingroup$ I guess, but we don’t get to know the population mean. After all, if we knew it, why would we go through the trouble of estimating it? // I’d also phrase it as “the sample mean is almost equal to E[X]” since the E[X] is a fixed value. $\endgroup$ – Dave Sep 10 '20 at 3:18
  • 1
    $\begingroup$ E[X] does not depend on the sample. Certainly $\bar{x}$ depends on the sample, however. $\endgroup$ – Dave Sep 10 '20 at 3:25
  • 1
    $\begingroup$ Yeah, and so whether the Sample Mean turns out to be same as E[X] is just a matter of chance of sample we took. They don't have to necessarily be the same always 👍 $\endgroup$ – Divyansh Singh Sep 10 '20 at 3:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.