The accompanying data file contains 40 observations on the response variable y along with the predictor variables x and d. Consider two linear regression models where Model 1 uses the variables x and d and Model 2 extends the model by including the interaction variable xd. Use the holdout method to compare the predictability of the models using the first 30 observations for training and the remaining 10 observations for validation. Click here for the Excel Data File a-1. Use the training set to estimate Models 1 and 2. Note: Negative values should be indicated by a minus sign. Round your answers to 2 decimal places. Predictor Variable Constant Model 1 (No interaction) Model 2 (Interaction) X d xd a-2. Calculate the RMSE of the two models in the validation set. Note: Do not round intermediate calculations and round final answers to 2 decimal places. Model 1 (No interaction) Model 2 (Interaction) RMSE a-3. Which model is better for making predictions? because its RMSE is 1 y A X B d 2 70 11 1 3 102 19 |1 476 12 1 5 83 14 1 6 61 17 10 7 62 13 0 8 67 20 0 9 98 16 1 10 84 11 1 11 101 15 1 12 51 16 0 13 108 16 1 14 32 13 0 15 71 15 1 16 101 17 1 17 90 15 1 18 112 19 1 19 88 13 |1 20 110 18 1 21 95 17 1 22 44 14 10 23 51 19 0 24 112 17 1 25 113 17 1 26 52 13 0 27 59 10 1 28 100 16 1 29 78 14 1 30 90 16 1 31 59 16 0 32 59 15 0 33 53 15 0 34 119 19 1 35 109 18 |1 36 68 11 0 37 104 19 1 38 45 18 0 39 67 17 0 40 65 15 0 41 74 14 1