Question 4a Let's use the 100th data point (X100, )100), i.e. k =100 to find these quantities! Complete the function simulate that takes in a data point xk , yk and return model_risk , model_var , and ratio , ratio of model variance to model risk. Here is a suggested format but you do not need to follow the skeleton code: . Assign predictions to a length 1000 vector where each element is the prediction on xk using a model in models . Note that sklearn's . predict return an array, but we only need a scaler prediction! Try to print out a prediction and ajust your code if needed. . Use predictions to compute model_risk . . Use predictions to compute model_var . . Use model_risk and model_var to compute the ratio , ratio of model variance to model risk. Hint: Use list comprehension for creating your predictions to save overhead of using multiple appends. def simulate (xk, yk, models) : x_100=mask_data. iloc[100,1:6] y_100=mask_data. iloc[ 100,0] predictions = [model. predict ([x_100]) [0] for model in models] model_risk = (y_100-predictions)**2 preds_df=pd. DataFrame ({ 'predicitons' :predictions, 'model_risk' :model_risk}) model_var = preds_df [ 'predictions' ]. var()/preds_df[ 'error_squared' ]. mean() ratio = model_var/model_risk return model_risk, model_var, ratio model_risk, model_var, ratio = simulate(x_100, y_100, models) model_risk, model_var, ratio