1. What factors influence a customer's decision to buy a car? What are the objectives of the model that Farid plans to build?
2. Explain (in words) the steps involved with building the neural network model with R programming language.
3. Build, validate, and interpret the neural network model using a different number of hidden neurons.
4. Build a linear regression model for comparison purposes. Discuss the advantages and disadvantages of your best neural network compared to your linear regression model.
5. Make a decision and offer your recommendation to Adam Ebraham.

1. 

When purchasing a car, people often gravitated toward different features. Some popular features included mctionalies like navigation systems, leather seats, seat heaters, and cmise con1rol. For cars intended for private use, personal preferences such as favourite colour and the desired vehicle shape came into play. Therefore, the price of one particular car model could vary according to these features. A car manufacturing plant had designed several new car models with different features and accessories. On June 11, ZlB, Adam Ebraham, the chief marketing ofcer at an automobile agency, was looking at a list, which he had received from the manufacturing plant, of diiferent car model features. After receiving the manufacturer's suggested retail price (MSRP), the dealerships would decide on a nal price that included environmental fees, air conditioning fees, and other fees. The MSRP was the base price of a vehicle model, set nationally for every dealer, before environmental costs and fees were applied. A car's features and accessories affected the MSRP. Ebraham needed to provide the MSRPs to dealers next week, and he had to decide on the base prices. He did not know how to determine which features were essential or to what extent each feature contributed to the price. He had heard of a prediction system called neural nehvorks that could help him to accurately predict the price. Yusuf Farid, a data scienst with expertise in machine learning and prediction, had recently been hired at the company research lab- Ebraham stopped by Farid's ofce and provided Farid with the data collected for the previous batch of cars (see Exhibit 1). There were 28 variables provided in the datasetone dependent variable and 2? independent variables. The dependent variable was the price of each car model. Farid hoped to provide Ebraham with an accurate MSRP recommendation for the new cars before the deadline the following week. EXHIBIT 1: CONFIGURATIONS OF THE PREVIOUS YEAR'S CARS Price Age KM Fuel HP MC Colour Auto CC Drs Cyl Grs Wight G_P Mfr_GABS Abag_1Abag_2AC Comp CD Clock Pwi PStr Radio SpM M_Rim Tow_Bar 21,000 26 31463 Petrol 195 0 Silver 0 1800 3 3 6 1189 10 1 0 1 0 1 O 20,000 23 43612 Petrol 195 0 Red 0 1800 3 3 1189 0 1 0 1 1 19,650 26 32191 Petrol 195 0 Red 0 1800 3 3 6 1189 0 21,550 32 23002 Petrol 195 1 Black 0 1800 3 3 6 1189 0 22,550 33 34133 Petrol 195 1 Grey 0 1800 3 3 6 1189 0 0 1 22,050 29 18741 Petrol 195 0 Grey 1800 3 3 6 1189 1 1 1 1 1 0 1 22,800 31 34002 Petrol 195 1 Grey 0 1800 3 3 5 1189 10 1 1 1 1 1 1 0 0 18,000 25 21718 Petrol 113 1 Blue 0 1600 3 3 5 1109 20 1 1 0 0 16,800 25 25565 Petrol 113 0 Grey 0 1600 3 3 5 1069 1 O KO 17,000 31 64361 Petrol 113 1 Grey 0 1600 3 3 5 1109 1 1 1 16,000 31 67662 Petrol 113 1 Blue 0 1600 3 3 5 1109 1 17,000 30 43907 Petrol 113 0 Grey 1 1600 3 3 5 1174 1 16,000 29 56351 Petrol 113 1 Black 0 1600 3 3 5 1124 1 1 17,000 29 32222 Petrol 113 1 Black 0 1600 3 3 5 1124 1 16,300 30 25815 Petrol 113 1 Grey 0 1600 3 3 5 1124 1 16,000 26 28452 Petrol 113 1 Blue 0 1600 3 3 5 1124 1 17,545 28 34547 Petrol 113 1 Blue 0 1600 3 3 5 1124 1 H 15,800 30 41417 Petrol 113 1 Black 0 1600 3 3 5 1124 1 17,000 29 44144 Petrol 113 0 Grey 0 1600 3 3 5 1124 1 18,000 31 11092 Petrol 113 1 Blue 0 1600 3 3 5 1124 1 1 13,000 30 9752 Petrol 100 1 Silver 0 1400 3 3 5 1104 10 1 1 15,800 23 35201 Petrol 100 1 Blue 0 1400 3 3 5 1104 1 16,000 28 29512 Petrol 100 1 Black 0 1400 3 3 5 1104 1 15,000 27 32694 Petrol 100 1 Grey 0 1400 3 3 5 1104 1 15,550 23 41002 Petrol 100 1 Grey 0 1400 3 3 5 1104 1 15,800 27 43002 Petrol 100 0 Grey 0 1400 3 3 5 1104 1 16,000 26 25002 Petrol 100 0 Grey 0 1400 3 3 5 1104 1 15,000 24 10002 Petrol 100 1 Blue 0 1400 3 3 5 1104 1 15,800 33 25331 Petrol 100 1 Green 1400 3 3 5 1104 1 19 14,800 28 27502 Petrol 100 0 Blue 0 1400 3 3 5 1104 3 1 14,000 23 49061 Petrol 100 0 Blue 0 1 0 1400 3 5 1104EXHIBIT 1 CONTINUED Configuration Definition Price Offer price in CA$ Age Age in months KM Accumulated kilometers on odometer Fuel Fuel type (Petrol, Diesel, CNG) HP Horse power MC Metallic colour? (Yes=1, No=0) Colour Colour (blue, red, grey, silver, black, etc.) Auto Automatic (Yes=1, No=0) CC Cylinder volume in cubic centimeters Drs Number of doors Cyl Number of cylinders Grs Number of gear positions Wght Weight in kilograms G P Guarantee period in months Mfr G Within manufacturer's guarantee period (Yes=1, No=0) ABS Anti-lock brake system (Yes=1, No=0) Abag_1 Driver airbag (Yes=1, No=0) Abag 2 Passenger airbag (Yes=1, No=0) AC Automatic air conditioning (Yes=1, No=0) Comp Board computer (Yes=1, No=0) CD CD player (Yes=1, No=0) Clock Central lock (Yes=1, No=0) Pwin Powered windows (Yes=1, No=0) PStr Power steering (Yes=1, No=0) Radio Radio (Yes=1, No=0) SPM Sport model (Yes=1, No=0) M Rim Metallic rim (Yes=1, No=0) Tow Bar Tow bar (Yes=1, No=0)