The key advantage of bivariate regression over correlation

You learned in this chapter that the key advantage of bivariate regression over correlation is that regression can be used for prediction. Explain this period how is it that regression can be used to predict values not in the data set, but correlation cannot?
Identify the two criteria that a dependent variable must meet ordinary least squares regression to be used?
Does OLS regression place restrictions on the levels of measurement of the independent variables?
Explain the advantage of multiple regression over bivariate regression. What does multiple regression do that bivariate regression does not? Why is this important?