Statistics 3022 Fall 2017 Regression project Due: Friday, November 3rd at 5:00 pm In this project, you are assigned a subset of variables from the data set FirstYearGPA, based on the last digit of your UMN Student ID number: If the last digit of When tting the models, When making predictions using your UMN Student the models, ID is... your assigned your assigned assume the predictors take the variable for binary variable for following values... Simple Linear Multiple Linear (Note: Use only the predictor(s) Regression is... Regression is... you have included in your nal Even (0, 2, 4, 6, 8) SATV = 670, HSGPA = 3.6, FirstGen = 1 Odd (1, 3, 5, 7, 9) ' SATM = 710, HSGPA = 3.7, White = 1 1. When you t and assess the model, answer the following questions: a) Simple Linear Regression model: Use {your assigned Simple Linear Regression variable} to predict the rst year GPA (GPA). Do the data need to be transformed? (3 points: 1 for creating model correctly, 1 for appropriate decision on transformation, 1 for checking assumptions) b) Multiple Linear Regression model: i. Use the step( ) mction (see Section 4.2 of the Lecture Notes) to create the best, simplest model using the three predictors {your assigned Simple Linear Regression variable}, High School GPA (HSGPA), and {your assigned Multiple Linear Regression binary variable}, as possibly the interaction terms, to predict the rst year GPA. Do NOT transform the response (GPA). (1 point for doing model selection) ii. Regardless of what model you chose in part 1. b) i. above, add the terms whether or not the student was male (Male) and the number of humanities credits (HU) into your model. Test to see if you need an interaction between HU and Male. (4 points: 1 for doing an Ftest, l for writing the hypotheses correctly, 1 for having a model with male*humanities, and 1 for having a model with male+hurnanities)