Let $X_1,\cdots,X_n\overset{iid}{\sim} F_X(x)$ be a random sample from a symmetric distribution with a defined mean. If need be, assume that $n$ is odd and that $F_X(x)$ is continuous.
Is it always the case that the empirical median is an unbiased estimator for the expected value?
Simulations suggest this is the case, and it certainly fits with my intuition, but then I go to prove it, remember the equation for the PDF of an order statistic, and do something else with my life. Perhaps I am missing a simple argument based on symmetry.
This seems to show that what I posit is true in the limit. What about in a finite sample?