d) Show or explain how the residual [error] sum of squares can be used to calculate s. e) Calculate the sum of squares of "docs" from the output above. f) Does the following graph suggest any violation of basic regression assumptions? Whether "yes" or "no", what assumptions are you considering? Histogram of the Residuals (response is lifeexpf) 14 Mean -2.78932E-14 StDev 6.994 N 121 12- 10- Frequency A 2- 0- -18 -12 -6 12 ResidualQuestion 4. The following regression results are drawn from a UN data set including 122 countries by regressing "lifeexpf" onto "docs". The variables included are: lifeexpf: female life expectancy in years [i.e., the average for all females in each country] docs: number of doctors per 10,000 of the population. Regression Analysis: lifeexpf versus docs The regression equation is lifeexpf = 58.2 + 0.783 docs Predictor Coef SE Coef T P Constant 58 . 2470 0. 8811 66.10 0. 000 docs 0. 78324 0. 05772 13.57 0.000 S = R-Sq = 60.78 R-Sq (adj ) = 60.4% Analysis of Variance Source DF SS MS F Regression 1 9082 .7 9082 . 7 184.14 Residual Error 119 5869.5 Total 120 14952.2 Please answer the following questions: a) Interpret the impact of docs on lifeexpf. b) Test whether the model can be used for predicting female life expectancy in a country [use a 95% confidence level] and show your work. c) Does the graph below [a plot of the residuals against fitted values] seem to contradict the answer to part b)? Please explain why or why not. Residuals Versus the Fitted Values (response is lifeexpf) 20- 10- Residual -10- -20- 60 65 70 75 80 85 90 95 Fitted Value