Regression Model
1.Does the distribution of ages matter in this case or is the average age all that is important? E.g., If two patient distributions have the same average age but have different shapes (say one is skewed left and the other is skewed right), will the average of the model predictions for each be the same?
*Hint try it for a small sample
2.John first took the averages of the independent variables and then applied the model when average age (and it’s square) changed to estimate average costs. Jane on the other hand, first applied the model to data with modified ages and then she took the average of estimated costs. Why does the averaging the independent variables before applying the model versus applying the model to the data and then averaging expected costs matter?
3.Whose analysis would you support and why?