The use of regression analysis to apply a mathematically sound method to determine if two variables are related in such a manner that one can predict the other. In fitting a regression line to the data, we will be able to make forecasts and projections based on knowledge we have of the existing data. Some questions as food for thought…
Do you suspect any correlated relationships in real-data sets?
Can you think of any relationships that would be positive (as one variable goes up, the other variable goes up)? What about negative relationships?
What is a reasonable level for the correlation value (for simple 2 variable linear regression and for multiple regression)? Why?
