Supposing that $X_1, X_2, ..., X_n$ are i.i.d with distribution $X_i \sim Bin(m_i,\theta) i=1,...,n$ where $m_i$ are positive integers. Consider the estimators of $\theta$

$T_1(X) = \frac{\sum_{i=1}^{n}X_i}{\sum_{i=1}^{n}m_i}$ and $T_2(X) = 1/n \sum_{i=1}^{n} \frac{X_i}{m_i}$

Prove that $ MSE(T_1) < MSE(T_2)$.

I have found that $ MSE(T_1) = \frac{\theta(1-\theta)}{\sum_{i=1}^{n}m_i}$ and $MSE(T_2) = \frac{\theta(1-\theta)}{n \sum_{i=1}^{n}1/m_i}$, but I couldn't prove that $\frac{n \sum_{i=1}^{n}1/m_i}{\sum_{i=1}^{n}m_i}$ is less than 1.

  • $\begingroup$ Welcome to Cross Validated! Please take a moment to view our tour. This looks like a self-study question. If that is the case, please tag your question appropriately and read the associated wiki. $\endgroup$ – Tavrock Apr 3 '17 at 17:47
  • $\begingroup$ The expression for $MSE(T_2)$ looks incorrect. $\endgroup$ – HStamper Apr 3 '17 at 18:08
  • $\begingroup$ Yes, You're right. Actually $MSE(T_2) = \frac{\theta(1-\theta)}{n} \sum_{i=1}^{n}1/m_i$ $\endgroup$ – Mayara Apr 3 '17 at 18:30
  • $\begingroup$ Correcting the expression of $MSE(T_2)$ I could prove this using that $(\sum a_i b_i)^2 \le (\sum a_i^2)(\sum b_i^2)$. Thanks for all help! $\endgroup$ – Mayara Apr 3 '17 at 18:33

Notice that $m_i\geq 1$. So:

$$n\sum_{i=1}^n\frac{1}{m_i}\leq n\leq\sum_{i=1}^n m_i.$$

| cite | improve this answer | |

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.