15.Which variable is the least useful for predicting Total Charge? Does this make practical sense? /3 a) Give a full explanation (form, direction, strength) 16) Which Month/Year is in fact the most expensive, relative to the fitted regression equation? /4 a) What might have happened in that month to create this outlier? Billing Days, Consumption (Cubic Meters). Average Temperature (degree 53% CI for Pearson Correlation lo I ... Method Correlation type Pearson Number of rows used 38 Correlations Consumption Average Date (Cubic Temper (DDMonYYYY) Billing Days Meters) (degre C )Degree Days Billing Days 0.032 Consumption (Cubic Meters) 0.329 -0.083 Average Temperature (degrees Ce -0 320 0.236 -0.897 Degree Days 0.320 -0.166 0.905 -0.996 Total Charges 0.176 -0.128 0.968 -0.924 0.930