## Step-by-Step Solution ### Step 1: Understanding the Cramér-Rao Lower BoundThe **Cramér-Rao lower bound (CR)** is a fundamental result in estimation theory. It provides a theoretical lower limit on the **variance** of any **unbiased estimator** of a parameter. ### Step 2: Statement of the CRLB For an unbiased estimator \(\hat{\theta}\) of a parameter \(\theta\): \[ \text{Var}(\hat{\theta}) \geq \frac{1}{I(\theta)} \] where \(I(\theta)\) is the **Fisher information**. ### Step 3: Analyzing the Options - **a) An upper bound for the variance of any unbiased estimator** **Incorrect.** The CRLB gives a lower bound, not an upper bound. - **b) A lower bound for the variance of any unbiased estimator** **Correct.** This is exactly what the CRLB provides. - **c) The exact variance of the maximum likelihood estimator** **Incorrect.** The CRLB gives a bound, not the exact variance. - **d) The asymptotic distribution of the MLE** **Incorrect.** The CRLB does not provide a distribution, just a variance bound. ### Step 4: Final Answer **The correct answer is:** \[ \boxed{\text{b) A lower bound for the variance of any unbiased estimator}} \] --- **Summary Table:** | Option | Description | Correct? | |--------|--------------------------------------------------|-----------| | a | Upper bound for variance | No | | b | Lower bound for variance | **Yes** | | c | Exact variance of MLE | No | | d | Asymptotic distribution of MLE | No |