Q2. (32 points) Let's continue with the same experimental data you have seen in Q1. Now we have some more problems to solve! (a). (8 points.) Consider the simple linear regression model fitted to the original data that was carried out in Q1 analysis, please calculate the joint 95% CI for Bo and B1 using the Bonferroni method. (b). (8 points.) Suppose Julliana is not a fan of transformation and she asked for other modeling choices that might be reasonable of this original data set. So you decided that our regression model is now of the form: Y; = 30 + B1X? + , where ; are independent and E(ei) = 0, V(@) = 02. Please rewrite the regression model using matrix approach, including the model assumptions. Define every vector or matrix you use in the model. (c). (10 points.) Please derive the least squares estimator of Bo and B, in this new model given in part (b) using the matrix approach, and show these are unbiased estimators. (d). (6 points.) Does the new model considered in (b) have a better fit compared to the simple linear regression model on the same original data set that was done in Q1? Why or why not? Please fit this new model on the data and use analysis findings to support your argument. Q2. (32 points) Let's continue with the same experimental data you have seen in Q1. Now we have some more problems to solve! (a). (8 points.) Consider the simple linear regression model fitted to the original data that was carried out in Q1 analysis, please calculate the joint 95% CI for Bo and B1 using the Bonferroni method. (b). (8 points.) Suppose Julliana is not a fan of transformation and she asked for other modeling choices that might be reasonable of this original data set. So you decided that our regression model is now of the form: Y; = 30 + B1X? + , where ; are independent and E(ei) = 0, V(@) = 02. Please rewrite the regression model using matrix approach, including the model assumptions. Define every vector or matrix you use in the model. (c). (10 points.) Please derive the least squares estimator of Bo and B, in this new model given in part (b) using the matrix approach, and show these are unbiased estimators. (d). (6 points.) Does the new model considered in (b) have a better fit compared to the simple linear regression model on the same original data set that was done in Q1? Why or why not? Please fit this new model on the data and use analysis findings to support your argument