Solve manually not using code
 

Exercise 1 (10 pts): We have collected the following data points that give the yearly medical expenses for a male person based on his age. We want to build a model that predicts the medical expenses for any male person of any age. SNUM AGE MEXP 1 15 100 2 20 135 3 25 135 4 37 150 5 40 250 6 45 270 7 481 290 8 50 360 9 55 275 10 61 400 11 64 500 12 67 1000 13 70 1500 We want to manually develop regression model (hint: do the calculations using Excel). a) What kind of a model is most suitable for the task? Explain (have proof). b) Compute the parameters of the linear regression. c) Compute the corresponding obtained cost function d) Compute the corresponding R2 score. e) Now use Gradient Descent, with a learning rate = 0.1. Compute the update of the parameters for three iterations of the GD.