Linear Regression Equation y = a + bx + b2x2 Where y is shoulder length, xy is weight and x2 is height. There are two steps involved to estimate the shoulder length: 1) Training process. The purpose of this process is to obtain a, b, and by in the equation. Assume there are total M customers in the file. We may choose 90% of them, sayn (=M*90%), for training. XL, *2 and y are md-column vectors. You are required to design your own experiments to complete the training process. cq.(1) (4, 14 E6-E 44% 1,1) Ex*XXXV-LEA, (24,-1X4,9% 4,4% ) 2. (EX XXX)-1X1,4, a = j-6,5, -b,5) eq.(2) eq.(3) whereyisthemeanaverage falueofnelementof y wherei, isthemear(average yalueofnelementof.x, wherex, isthemeanta verage kalueofnelementsfx 2) Estimating/predicting process. The purpose of this process is to predict y for the customers whose data were not used for training and to validate y for the customers whose data were used in the training process, During the process, y is calculated through the equation below: y = a + bx + b2x2 where a, b, and b are known parameters obtained in the training process. You are required to design your own experiments to complete the predicting process. To evaluate the performance of the estimation, we define the accuracy rate as follows: accuracy rate = Yr-Yratisnase!. 100 eq. (4) where yestinoce is the shoulder length that is obtained in the estimating process, Vreer is the shoulder length in the file "customerdata.xlsx". Problem 1. Design/develop a function named "train.m" for the training process that will obtain a, b, and b). The function need read the file "customerdata.xlsx" for x1, x and y. Your script should prompt the user for the percentage of customers who are used for the training. Dr. Sophie Liu, Engineering Department Linear Regression Equation y = a + bx + b2x2 Where y is shoulder length, xy is weight and x2 is height. There are two steps involved to estimate the shoulder length: 1) Training process. The purpose of this process is to obtain a, b, and by in the equation. Assume there are total M customers in the file. We may choose 90% of them, sayn (=M*90%), for training. XL, *2 and y are md-column vectors. You are required to design your own experiments to complete the training process. cq.(1) (4, 14 E6-E 44% 1,1) Ex*XXXV-LEA, (24,-1X4,9% 4,4% ) 2. (EX XXX)-1X1,4, a = j-6,5, -b,5) eq.(2) eq.(3) whereyisthemeanaverage falueofnelementof y wherei, isthemear(average yalueofnelementof.x, wherex, isthemeanta verage kalueofnelementsfx 2) Estimating/predicting process. The purpose of this process is to predict y for the customers whose data were not used for training and to validate y for the customers whose data were used in the training process, During the process, y is calculated through the equation below: y = a + bx + b2x2 where a, b, and b are known parameters obtained in the training process. You are required to design your own experiments to complete the predicting process. To evaluate the performance of the estimation, we define the accuracy rate as follows: accuracy rate = Yr-Yratisnase!. 100 eq. (4) where yestinoce is the shoulder length that is obtained in the estimating process, Vreer is the shoulder length in the file "customerdata.xlsx". Problem 1. Design/develop a function named "train.m" for the training process that will obtain a, b, and b). The function need read the file "customerdata.xlsx" for x1, x and y. Your script should prompt the user for the percentage of customers who are used for the training. Dr. Sophie Liu, Engineering Department