In situations where expected rate on invested differ substantially, standard deviation may not provide a good picture of one investment's stand-alone risk relative to another. In this situation what other measure of risk would you use? How do you define that measure?