Source SS df MS Number of obs = F(1, 980) 982 = 5.06 Model Residual 1.0822e+11 2.0961e+13 1 1.0822e+11 980 2.1389e+10 Prob F R-squared = 0.0247 = 0.0051 Total 2.1070e+13 981 2.1478e+10 Adji R-squared = Root MSE 0.0041 1.5e+05 revenue_from_c~t Coef. Std. Err. t P>|t| [95% Conf. Interval] tenure_of_client _cons 71.79758 31.91979 51501.72 6581.747 2.25 7.82 0.025 9.158584 134.4366 0.000 38585.78 64417.66 3. Consider now the following output, which is from the same dataset as the one above, except this time, in addition to the previous model, a binary explanatory variable has been introduced (custtype 1). This variable is equal to 1 if the customer may be considered as a "loyal customer", and to 0, if the customer is a "new or not too loyal" customer. Source SS df MS Model Residual 3.4672e+11 2 1.7336e+11 979 2.1167e+10 Number of obs F(2, 979) Prob > F R-squared = 982 = 8.19 = 0.0003 = Total 2.1070e+13 981 2.1478e+10 Adj R-squared Root MSE = = 1.5e+05 revenue_from_c~t tenure_of_client Coef. Std. Err. t P>|t| [95% Conf. Interval] custtype1 _ cons 48.10095 32.52914 32117.58 9568.18 40621.69 -15.73392 111.9358 3.36 0.001 5.56 0.000 26284.67 54958.71 a) Please calculate all of the blanked out numbers and explain in much detail how you have gotten to them. Note that as per scientific notation, the term "e+X" should be read as "*10x". b) Would an F-test justify the inclusion of this additional variable? Please perform the F-test and justify in detail. Incollage