Your assignment is to use regression techniques to determine factors that predict player salary. I would suggest the following steps : a- Make a matrix plot of continuous variables to determine which variables look like they might be good predictors of salary, and conrm that you should make a log transformation of salary (either plots or use Box-Cox]. Look for potential outliers and variables that need transformation. ALSO _ make boxplots of loglsalary} vs. team and vs. position (which are categorical predictors}. Comment on what you observe. After you show why you should use log(sa|ary}, redo matrix plot with log(salary} as the rst variable. 0 Fit a series of regression models to find which variables are the best predictors of Ioglsalary). You probably won't be able to use best'subsets regression since MINITAB can't handle more than 31 variables, and team, for example, requires 15 indicator variables. Better will be backwards stepwise regression. Write a few sentences about what you did, and provides results for your first and last models, whatever they might be. c Having chosen a final model, look at residual plots to check model fit and look for any problems. 0 Write a concise paragraph in which you discuss the meaning and interpretation of the model coefficients, the significance of various terms, the overall model t, etc. Be sure to discuss if any outliers or influential points seem to be present. Including relevant plots and MINITAB/SPSS/R output, discuss your findings, identify significant predictors, and outline your conclusions