library(AER) library(MASS) data(Boston) # run a regression of medv on all remaining variables in the Boston data set # obtain a robust summary of the coefficients # what is the R^2 of the model? # run a regression of medv on all remaining variables in the Boston data set full_mod <- lm(medv ~., data = Boston) # obtain a robust summary of the coefficients coeftest(full_mod, vcov. = vcovHC, type = "HC1") # what is the R^2 of the model? summary(full_mod)$adj.r.squared test_object("full_mod") test_function("coeftest", args="x") test_student_typed("vcov. = vcovHC") success_msg("Right. Notice that the smallest p-value is associated with the coefficient on ptratio, the pupil-teacher ratio by town. It is conceivable that quality of the schooling district is an important location factor. According to the adj. R^2 , the full model does better than the model dealt with in exercises 2 and 3 which uses a smaller subset of the variables as regressors.")