Model the amount of money (in pounds) spent gambling by British teens (in the 1980s) using the other variables in the teengamb data set from the faraway package as predictors. Which model do you like best? Be sure to investigate possible interactions between sex and other predictors.