Two panels show MSE and coefficient values as a function of $$lambda$,$ based on k$-$fold cross validation of models with different values of $$lambda$.$ The lines have been identified with colour$-$coded labels. Line F $($which goes to zero around $$lambda subscript $1$ S E end subscript$)$ is the coefficient of zipcode. Can you explain why there is a bend in the line representing the coefficient of long when line F $($yellow$)$ goes to zero? Choose all that apply.

A$)$ Zipcode makes a bigger contribution to the model than longitude.

B$)$ Property taxes in different zip codes have an impact on house prices.

C$)$ Lasso regression explicitly connects the values of different coefficients through its cost function.

D$)$ Longitude and zipcode violate the assumption of independent predictor variables.