I was wondering if the the following is an example of the law of iterated expectations.

Say we observe the entire population of the random variable $X$. Call the mean of the population $\mu$ . Now, let us arbitrarily resample a fraction of this population n times. For each of these samples, obtain an unbiased estimate of the mean as $\hat{\mu_{i}}$ where i denotes the sample. Now, we can obtain the average of these sample averages as:$$\tilde{\mu}=\frac{1}{N}\sum_{i}\hat{\mu_{i}}$$ Given that our estimator is unbiased, it should be that $\tilde{\mu}$ is extremely close to $\mu$ . Now, can this also be thought of as the Law of Iterated Expectations where $$E[X]=E[E[X|i]]$$


1 Answer 1


This appears to be a trivial application of the Law of Iterated Expectations, because $X$ is mean-independent of the choice of the sample that contains realizations of it (i.e. the true mean of $X$ does not change as we change the sample). So

$$E[E(X|i)] = E[E(X)] = E(X)$$

The expression is not wrong, but it does not appear to me to be a useful example for the application of the tower property, since the result comes about due to independence, rather than from "averaging".


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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