Assume we have an estimator $\bar{\theta}$ for a parameter $\theta$. The bias of the estimator is the difference between the true value of the estimator, and its expected value:

$$\operatorname{Bias}(\bar{\theta}) = \operatorname{E}(\bar{\theta} - \theta)$$

If $\operatorname{Bias}(\bar{\theta})= 0$, then $\bar{\theta}$ is called unbiased. The variance of the estimator $\bar{\theta}$ is the expected square deviation from its expected value:

$$\operatorname{var}(\bar{\theta}) = \operatorname{E}[(\bar{\theta} − E[\bar{\theta} ])^2]$$

The mean squared error of the estimator $\bar{\theta}$ is

$$\operatorname{Error}(\bar{\theta}) = \operatorname{E}(\bar{\theta} − \theta)^2 = \operatorname{Bias}(\bar{\theta})^2+\operatorname{var}(\bar{\theta})$$

Let $X_1, \ldots , X_N$ be a sample of i.i.d random variables. Assume that $X_i$ has mean $\mu$ and variance $\sigma^2$ . Calculate the bias, variance and mean squared error of the following mean estimators:

(a) $\bar{\mu} =\frac 1 N \sum_{i=1}^N X_i $(i.e. the sample mean)

Bias I calculated as Below

\begin{align} & \operatorname{E}(\bar{\mu})-\mu \\[10pt] = {} & \operatorname{E} \left( \frac 1 N \sum_{i=1}^N X_i \right) - \mu \\[10pt] = {} & \frac 1 N \sum_{i=1}^N \operatorname{E}(X_i) - \mu \\[10pt] = {} & \frac N N \mu - \mu \\[10pt] \operatorname{Bias}(\bar{\mu}) = {} & 0 \end{align}

But which identity i need to use for calculating the variance??

  • 1
    $\begingroup$ You are missing the self-study tag to indicate the learning side of your question and avoid closure. $\endgroup$ – Xi'an Jan 11 '17 at 21:29

$$ \operatorname{var}\left( \frac 1 N \sum_{i=1}^N X_i\right) = \frac 1 {N^2} \operatorname{var}\left( \sum_{i=1}^N X_i \right) = \frac 1 {N^2} \sum_{i=1}^N \operatorname{var}(X_i) = \cdots $$

  • 1
    $\begingroup$ typo to fix in the first term... $\endgroup$ – Xi'an Jan 11 '17 at 21:30

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.