We have $X_1,...,X_n$ as $n$ independent random variables under the Bernoulli distribution i.e.:

$$P(X_i=1)=p$$ $$P(X_i=0)=1-p$$

where, $p$ is an unknown parameter. The distribution $Y=\sum_1^n X_i$ is the binomial distribution such that $Y\sim \mathrm{Binom}(n,p)$. I have found that $\bar X$ is an estimate of $Y/n$ but now it asks to find $E(\bar X(1-\bar X))$ as $(n-1)p(1-p)/n$

Any ideas how to calculate this the best way, would I expand or separate to $\mathrm{E}(\bar X)\mathrm{E}(1-\bar X)$? Would this involve manipulating the binomial formula?


Hint: $$ E[\bar{X}(1-\bar{X})]=E[\bar{X}]-E[\bar{X}^2]=\frac{1}{n}E[Y]-\frac{1}{n^2}E[Y^2] $$


Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.