One of the reasons why the normal distribution is so common is that It is the best distribution to describe data that only takes on two values It works well with the interquartile range It naturally arises when the data combines fluctuations across time It naturally arises when the forces shaping the data systematically increase in strength Question 22 Say that you have two variables, X and Y. You know they have a positive correlation. The mean of the conditional distribution of Y when X=10 is zero. What would you expect the mean of the conditional distribution to be when X=20 (i.e. something bigger than 0 )?