Suppose that X and Y are independent, identically distributed random variables with mean 0 and variance 1, such that
Show that X and Y are standard normal.
If we know that X and Y are standard normal, then it is easy to verify the relation. The idea thus is to prove the converse.